Tag Archives: Research

Chain-of-table: Evolving tables in the reasoning chain for table understanding

Posted by Zilong Wang, Student Researcher, and Chen-Yu Lee, Research Scientist, Cloud AI Team

People use tables every day to organize and interpret complex information in a structured, easily accessible format. Due to the ubiquity of such tables, reasoning over tabular data has long been a central topic in natural language processing (NLP). Researchers in this field have aimed to leverage language models to help users answer questions, verify statements, and analyze data based on tables. However, language models are trained over large amounts of plain text, so the inherently structured nature of tabular data can be difficult for language models to fully comprehend and utilize.

Recently, large language models (LLMs) have achieved outstanding performance across diverse natural language understanding (NLU) tasks by generating reliable reasoning chains, as shown in works like Chain-of-Thought and Least-to-Most. However, the most suitable way for LLMs to reason over tabular data remains an open question.

In “Chain-of-Table: Evolving Tables in the Reasoning Chain for Table Understanding”, we propose a framework to tackle table understanding tasks, where we train LLMs to outline their reasoning step by step, updating a given table iteratively to reflect each part of a thought process, akin to how people solve the table-based problems. This enables the LLM to transform the table into simpler and more manageable segments so that it can understand and analyze each part of the table in depth. This approach has yielded significant improvements and achieved new state-of-the-art results on the WikiTQ, TabFact, and FeTaQA benchmarks. The figure below shows the high-level overview of the proposed Chain-of-Table and other methods.

Given a complex table where a cyclist’s nationality and name are in the same cell, (a) generic, multi-step reasoning is unable to provide the correct answer (b) program-aided reasoning generates and executes programs (e.g., SQL queries) to deliver the answer, but falls short in accurately addressing the question. In contrast, (c) Chain-of-Table iteratively samples a chain of operations that effectively transform the complex table into a version specifically tailored to the question.

Chain-of-Table

In Chain-of-Table, we guide LLMs using in-context learning to iteratively generate operations and to update the table to represent its reasoning chain over tabular data. This enables LLMs to dynamically plan the next operation based on the results of previous ones. This continuous evolution of the table forms a chain, which provides a more structured and clear representation of the reasoning process for a given problem and enables more accurate and reliable predictions from the LLM.

For example, when asked, “Which actor has the most NAACP image awards?” the Chain-of-Table framework prompts an LLM to generate tabular operations mirroring tabular reasoning processes. It first identifies the relevant columns. Then, it aggregates rows based on shared content. Finally, it reorders the aggregated results to yield a final table that clearly answers the posed question.

These operations transform the table to align with the question presented. To balance performance with computational expense on large tables, we construct the operation chain according to a subset of tabular rows.. Meanwhile, the step-by-step operations reveal the underlying reasoning process through the display of intermediate results from the tabular operations, fostering enhanced interpretability and understanding.

Illustration of the tabular reasoning process in Chain-of-Table. This iterative process involves dynamically planning an operation chain and accurately storing intermediate results in the transformed tables. These intermediate tables serve as a tabular thought process that can guide the LLM to land to the correct answer more reliably.

Chain-of-Table consists of three main stages. In the first stage, it instructs the LLM to dynamically plan the next operation by in-context learning. Specifically, the prompt involves three components as shown in the following figure:

  1. The question Q: “Which country had the most cyclists finish in the top 3?”
  2. The operation history chain: f_add_col(Country) and f_select_row(1, 2, 3).
  3. The latest intermediate table T: the transformed intermediate table.

By providing the triplet (T, Q, chain) in the prompt, the LLM can observe the previous tabular reasoning process and select the next operation from the operation pool to complete the reasoning chain step by step.

Illustration of how Chain-of-Table selects the next operation from the operation pool and generates the arguments for the operation.(a) Chain-of-Table samples the next operation from the operation pool. (b) It takes the selected operation as input and generates its arguments.

After the next operation f is determined, in the second stage, we need to generate the arguments. As above, Chain-of-Table considers three components in the prompt as shown in the figure: (1) the question, (2) the selected operation and its required arguments, and (3) the latest intermediate table.

For instance, when the operation f_group_by is selected, it requires a header name as its argument.

The LLM selects a suitable header within the table. Equipped with the selected operation and the generated arguments, Chain-of-Table executes the operation and constructs a new intermediate table for the following reasoning.

Chain-of-Table iterates the previous two stages to plan the next operation and generate the required arguments. During this process, we create an operation chain acting as a proxy for the tabular reasoning steps. These operations generate intermediate tables presenting the results of each step to the LLM. Consequently, the output table contains comprehensive information about the intermediate phases of tabular reasoning. In our final stage, we employ this output table in formulating the final query and prompt the LLM along with the question for the final answer.


Experimental setup

We use PaLM 2-S and GPT 3.5 as the backbone LLMs and conduct the experiments on three public table understanding benchmarks: WikiTQ, TabFact, and FeTaQA. WikiTQ and FeTaQA are datasets for table-based question answering. TabFact is a table-based fact verification benchmark. In this blogpost, we will focus on the results on WikiTQ and TabFact. We compare Chain-of-Table with the generic reasoning methods (e.g., End-to-End QA, Few-Shot QA, and Chain-of-Thought) and the program-aided methods (e.g., Text-to-SQL, Binder, and Dater).


More accurate answers

Compared to the generic reasoning methods and program-aided reasoning methods, Chain-of-Table achieves better performance across PaLM 2 and GPT 3.5. This is attributed to the dynamically sampled operations and the informative intermediate tables.

Understanding results on WikiTQ and TabFact with PaLM 2 and GPT 3.5 compared with various models.

Better robustness on harder questions

In Chain-of-Table, longer operation chains indicate the higher difficulty and complexity of the questions and their corresponding tables. We categorize the test samples according to their operation lengths in Chain-of-Table. We compare Chain-of-Table with Chain-of-Thought and Dater, as representative generic and program-aided reasoning methods. We illustrate this using results from PaLM 2 on WikiTQ.

Performance of Chain-of-Thought, Dater, and the proposed Chain-of-Table on WikiTQ for questions that require an operation chain of varying lengths. Our proposed atomic operations significantly improve performance over generic and program-aided reasoning counterparts.

Notably, Chain-of-Table consistently surpasses both baseline methods across all operation chain lengths, with a significant margin up to 11.6% compared with Chain-of-Thought, and up to 7.9% compared with Dater. Moreover, the performance of Chain-of-Table declines gracefully with increasing number of operations compared to other baseline methods, exhibiting only a minimal decrease when the number of operations increases from four to five.


Better robustness with larger tables

We categorize the tables from WikiTQ into three groups based on token number: small (<2000 tokens), medium (2000 to 4000 tokens) and large (>4000 tokens). We then compare Chain-of-Table with Dater and Binder, the two latest and strongest baselines.

Performance of Binder, Dater, and the proposed Chain-of-Table on small (<2000 tokens), medium (2000 to 4000 tokens), and large (>4000 tokens) tables from WikiTQ. We observe that the performance decreases with larger input tables while Chain-of-Table diminishes gracefully, achieving significant improvements over competing methods. (As above, underlined text denotes the second-best performance; bold denotes the best performance.)

Performance of Binder, Dater, and the proposed Chain-of-Table on small (<2000 tokens), medium (2000 to 4000 tokens), and large (>4000 tokens) tables from WikiTQ. We observe that the performance decreases with larger input tables while Chain-of-Table diminishes gracefully, achieving significant improvements over competing methods. (As above, underlined text denotes the second-best performance; bold denotes the best performance.)

As anticipated, the performance decreases with larger input tables, as models are required to reason through longer contexts. Nevertheless, the performance of the proposed Chain-of-Table diminishes gracefully, achieving a significant 10+% improvement over the second best competing method when dealing with large tables. This demonstrates the efficacy of the reasoning chain in handling long tabular inputs.


Conclusion

Our proposed Chain-of-Table method enhances the reasoning capability of LLMs by leveraging the tabular structure to express intermediate steps for table-based reasoning. It instructs LLMs to dynamically plan an operation chain according to the input table and its associated question. This evolving table design sheds new light on the understanding of prompting LLMs for table understanding.


Acknowledgements

This research was conducted by Zilong Wang, Hao Zhang, Chun-Liang Li, Julian Martin Eisenschlos, Vincent Perot, Zifeng Wang, Lesly Miculicich, Yasuhisa Fujii, Jingbo Shang, Chen-Yu Lee, Tomas Pfister. Thanks to Chih-Kuan Yeh and Sergey Ioffe for their valuable feedback.

Source: Google AI Blog


Health-specific embedding tools for dermatology and pathology

Posted by Dave Steiner, Clinical Research Scientist, Google Health, and Rory Pilgrim, Product Manager Google Research

There’s a worldwide shortage of access to medical imaging expert interpretation across specialties including radiology, dermatology and pathology. Machine learning (ML) technology can help ease this burden by powering tools that enable doctors to interpret these images more accurately and efficiently. However, the development and implementation of such ML tools are often limited by the availability of high-quality data, ML expertise, and computational resources.

One way to catalyze the use of ML for medical imaging is via domain-specific models that utilize deep learning (DL) to capture the information in medical images as compressed numerical vectors (called embeddings). These embeddings represent a type of pre-learned understanding of the important features in an image. Identifying patterns in the embeddings reduces the amount of data, expertise, and compute needed to train performant models as compared to working with high-dimensional data, such as images, directly. Indeed, these embeddings can be used to perform a variety of downstream tasks within the specialized domain (see animated graphic below). This framework of leveraging pre-learned understanding to solve related tasks is similar to that of a seasoned guitar player quickly learning a new song by ear. Because the guitar player has already built up a foundation of skill and understanding, they can quickly pick up the patterns and groove of a new song.

Path Foundation is used to convert a small dataset of (image, label) pairs into (embedding, label) pairs. These pairs can then be used to train a task-specific classifier using a linear probe, (i.e., a lightweight linear classifier) as represented in this graphic, or other types of models using the embeddings as input.

Once the linear probe is trained, it can be used to make predictions on embeddings from new images. These predictions can be compared to ground truth information in order to evaluate the linear probe's performance.

In order to make this type of embedding model available and drive further development of ML tools in medical imaging, we are excited to release two domain-specific tools for research use: Derm Foundation and Path Foundation. This follows on the strong response we’ve already received from researchers using the CXR Foundation embedding tool for chest radiographs and represents a portion of our expanding research offerings across multiple medical-specialized modalities. These embedding tools take an image as input and produce a numerical vector (the embedding) that is specialized to the domains of dermatology and digital pathology images, respectively. By running a dataset of chest X-ray, dermatology, or pathology images through the respective embedding tool, researchers can obtain embeddings for their own images, and use these embeddings to quickly develop new models for their applications.


Path Foundation

In “Domain-specific optimization and diverse evaluation of self-supervised models for histopathology”, we showed that self-supervised learning (SSL) models for pathology images outperform traditional pre-training approaches and enable efficient training of classifiers for downstream tasks. This effort focused on hematoxylin and eosin (H&E) stained slides, the principal tissue stain in diagnostic pathology that enables pathologists to visualize cellular features under a microscope. The performance of linear classifiers trained using the output of the SSL models matched that of prior DL models trained on orders of magnitude more labeled data.

Due to substantial differences between digital pathology images and “natural image” photos, this work involved several pathology-specific optimizations during model training. One key element is that whole-slide images (WSIs) in pathology can be 100,000 pixels across (thousands of times larger than typical smartphone photos) and are analyzed by experts at multiple magnifications (zoom levels). As such, the WSIs are typically broken down into smaller tiles or patches for computer vision and DL applications. The resulting images are information dense with cells or tissue structures distributed throughout the frame instead of having distinct semantic objects or foreground vs. background variations, thus creating unique challenges for robust SSL and feature extraction. Additionally, physical (e.g., cutting) and chemical (e.g., fixing and staining) processes used to prepare the samples can influence image appearance dramatically.

Taking these important aspects into consideration, pathology-specific SSL optimizations included helping the model learn stain-agnostic features, generalizing the model to patches from multiple magnifications, augmenting the data to mimic scanning and image post processing, and custom data balancing to improve input heterogeneity for SSL training. These approaches were extensively evaluated using a broad set of benchmark tasks involving 17 different tissue types over 12 different tasks.

Utilizing the vision transformer (ViT-S/16) architecture, Path Foundation was selected as the best performing model from the optimization and evaluation process described above (and illustrated in the figure below). This model thus provides an important balance between performance and model size to enable valuable and scalable use in generating embeddings over the many individual image patches of large pathology WSIs.

SSL training with pathology-specific optimizations for Path Foundation.

The value of domain-specific image representations can also be seen in the figure below, which shows the linear probing performance improvement of Path Foundation (as measured by AUROC) compared to traditional pre-training on natural images (ImageNet-21k). This includes evaluation for tasks such as metastatic breast cancer detection in lymph nodes, prostate cancer grading, and breast cancer grading, among others.

Path Foundation embeddings significantly outperform traditional ImageNet embeddings as evaluated by linear probing across multiple evaluation tasks in histopathology.


Derm Foundation

Derm Foundation is an embedding tool derived from our research in applying DL to interpret images of dermatology conditions and includes our recent work that adds improvements to generalize better to new datasets. Due to its dermatology-specific pre-training it has a latent understanding of features present in images of skin conditions and can be used to quickly develop models to classify skin conditions. The model underlying the API is a BiT ResNet-101x3 trained in two stages. The first pre-training stage uses contrastive learning, similar to ConVIRT, to train on a large number of image-text pairs from the internet. In the second stage, the image component of this pre-trained model is then fine-tuned for condition classification using clinical datasets, such as those from teledermatology services.

Unlike histopathology images, dermatology images more closely resemble the real-world images used to train many of today's computer vision models. However, for specialized dermatology tasks, creating a high-quality model may still require a large dataset. With Derm Foundation, researchers can use their own smaller dataset to retrieve domain-specific embeddings, and use those to build smaller models (e.g., linear classifiers or other small non-linear models) that enable them to validate their research or product ideas. To evaluate this approach, we trained models on a downstream task using teledermatology data. Model training involved varying dataset sizes (12.5%, 25%, 50%, 100%) to compare embedding-based linear classifiers against fine-tuning.

The modeling variants considered were:

  • A linear classifier on frozen embeddings from BiT-M (a standard pre-trained image model)
  • Fine-tuned version of BiT-M with an extra dense layer for the downstream task
  • A linear classifier on frozen embeddings from the Derm Foundation API
  • Fine-tuned version of the model underlying the Derm Foundation API with an extra layer for the downstream task

We found that models built on top of the Derm Foundation embeddings for dermatology-related tasks achieved significantly higher quality than those built solely on embeddings or fine tuned from BiT-M. This advantage was found to be most pronounced for smaller training dataset sizes.

These results demonstrate that the Derm Foundation tooI can serve as a useful starting point to accelerate skin-related modeling tasks. We aim to enable other researchers to build on the underlying features and representations of dermatology that the model has learned.

However, there are limitations with this analysis. We're still exploring how well these embeddings generalize across task types, patient populations, and image settings. Downstream models built using Derm Foundation still require careful evaluation to understand their expected performance in the intended setting.


Access Path and Derm Foundation

We envision that the Derm Foundation and Path Foundation embedding tools will enable a range of use cases, including efficient development of models for diagnostic tasks, quality assurance and pre-analytical workflow improvements, image indexing and curation, and biomarker discovery and validation. We are releasing both tools to the research community so they can explore the utility of the embeddings for their own dermatology and pathology data.

To get access, please sign up to each tool's terms of service using the following Google Forms.

After gaining access to each tool, you can use the API to retrieve embeddings from dermatology images or digital pathology images stored in Google Cloud. Approved users who are just curious to see the model and embeddings in action can use the provided example Colab notebooks to train models using public data for classifying six common skin conditions or identifying tumors in histopathology patches. We look forward to seeing the range of use-cases these tools can unlock.


Acknowledgements

We would like to thank the many collaborators who helped make this work possible including Yun Liu, Can Kirmizi, Fereshteh Mahvar, Bram Sterling, Arman Tajback, Kenneth Philbrik, Arnav Agharwal, Aurora Cheung, Andrew Sellergren, Boris Babenko, Basil Mustafa, Jan Freyberg, Terry Spitz, Yuan Liu, Pinal Bavishi, Ayush Jain, Amit Talreja, Rajeev Rikhye, Abbi Ward, Jeremy Lai, Faruk Ahmed, Supriya Vijay,Tiam Jaroensri, Jessica Loo, Saurabh Vyawahare, Saloni Agarwal, Ellery Wulczyn, Jonathan Krause, Fayaz Jamil, Tom Small, Annisah Um'rani, Lauren Winer, Sami Lachgar, Yossi Matias, Greg Corrado, and Dale Webster.

Source: Google AI Blog


Social learning: Collaborative learning with large language models

Posted by Amirkeivan Mohtashami, Research Intern, and Florian Hartmann, Software Engineer, Google Research

Large language models (LLMs) have significantly improved the state of the art for solving tasks specified using natural language, often reaching performance close to that of people. As these models increasingly enable assistive agents, it could be beneficial for them to learn effectively from each other, much like people do in social settings, which would allow LLM-based agents to improve each other’s performance.

To discuss the learning processes of humans, Bandura and Walters described the concept of social learning in 1977, outlining different models of observational learning used by people. One common method of learning from others is through a verbal instruction (e.g., from a teacher) that describes how to engage in a particular behavior. Alternatively, learning can happen through a live model by mimicking a live example of the behavior.

Given the success of LLMs mimicking human communication, in our paper “Social Learning: Towards Collaborative Learning with Large Language Models”, we investigate whether LLMs are able to learn from each other using social learning. To this end, we outline a framework for social learning in which LLMs share knowledge with each other in a privacy-aware manner using natural language. We evaluate the effectiveness of our framework on various datasets, and propose quantitative methods that measure privacy in this setting. In contrast to previous approaches to collaborative learning, such as common federated learning approaches that often rely on gradients, in our framework, agents teach each other purely using natural language.


Social learning for LLMs

To extend social learning to language models, we consider the scenario where a student LLM should learn to solve a task from multiple teacher entities that already know that task. In our paper, we evaluate the student’s performance on a variety of tasks, such as spam detection in short text messages (SMS), solving grade school math problems, and answering questions based on a given text.

A visualization of the social learning process: A teacher model provides instructions or few-shot examples to a student model without sharing its private data.

Language models have shown a remarkable capacity to perform tasks given only a handful of examples–a process called few-shot learning. With this in mind, we provide human-labeled examples of a task that enables the teacher model to teach it to a student. One of the main use cases of social learning arises when these examples cannot be directly shared with the student due, for example, to privacy concerns.

To illustrate this, let’s look at a hypothetical example for a spam detection task. A teacher model is located on device where some users volunteer to mark incoming messages they receive as either “spam” or “not spam”. This is useful data that could help train a student model to differentiate between spam and not spam, but sharing personal messages with other users is a breach of privacy and should be avoided. To prevent this, a social learning process can transfer the knowledge from the teacher model to the student so it learns what spam messages look like without needing to share the user’s personal text messages.

We investigate the effectiveness of this social learning approach by analogy with the established human social learning theory that we discussed above. In these experiments, we use PaLM 2-S models for both the teacher and the student.

A systems view of social learning: At training time, multiple teachers teach the student. At inference time, the student is using what it learned from the teachers.

Synthetic examples

As a counterpart to the live teaching model described for traditional social learning, we propose a learning method where the teachers generate new synthetic examples for the task and share them with the student. This is motivated by the idea that one can create a new example that is sufficiently different from the original one, but is just as educational. Indeed, we observe that our generated examples are sufficiently different from the real ones to preserve privacy while still enabling performance comparable to that achieved using the original examples.

The 8 generated examples perform as well as the original data for several tasks (see our paper).

We evaluate the efficacy of learning through synthetic examples on our task suite. Especially when the number of examples is high enough, e.g., n = 16, we observe no statistically significant difference between sharing original data and teaching with synthesized data via social learning for the majority of tasks, indicating that the privacy improvement does not have to come at the cost of model quality.

Generating 16 instead of just 8 examples further reduces the performance gap relative to the original examples.

The one exception is spam detection, for which teaching with synthesized data yields lower accuracy. This may be because the training procedure of current models makes them biased to only generate non-spam examples. In the paper, we additionally look into aggregation methods for selecting good subsets of examples to use.


Synthetic instruction

Given the success of language models in following instructions, the verbal instruction model can also be naturally adapted to language models by having the teachers generate an instruction for the task. Our experiments show that providing such a generated instruction effectively improves performance over zero-shot prompting, reaching accuracies comparable to few-shot prompting with original examples. However, we did find that the teacher model may fail on certain tasks to provide a good instruction, for example due to a complicated formatting requirement of the output.

For Lambada, GSM8k, and Random Insertion, providing synthetic examples performs better than providing generated instructions, whereas in the other tasks generated instruction obtains a higher accuracy. This observation suggests that the choice of the teaching model depends on the task at hand, similar to how the most effective method for teaching people varies by task.

Depending on the task, generating instructions can work better than generating new examples.

Memorization of the private examples

We want teachers in social learning to teach the student without revealing specifics from the original data. To quantify how prone this process is to leaking information, we used Secret Sharer, a popular method for quantifying to what extent a model memorizes its training data, and adapted it to the social learning setting. We picked this method since it had previously been used for evaluating memorization in federated learning.

To apply the Secret Sharer method to social learning, we design “canary” data points such that we can concretely measure how much the training process memorized them. These data points are included in the datasets used by teachers to generate new examples. After the social learning process completes, we can then measure how much more confident the student is in the secret data points the teacher used, compared to similar ones that were not shared even with the teachers.

In our analysis, discussed in detail in the paper, we use canary examples that include names and codes. Our results show that the student is only slightly more confident in the canaries the teacher used. In contrast, when the original data points are directly shared with the student, the confidence in the included canaries is much higher than in the held-out set. This supports the conclusion that the teacher does indeed use its data to teach without simply copying it over.


Conclusion and next steps

We introduced a framework for social learning that allows language models with access to private data to transfer knowledge through textual communication while maintaining the privacy of that data. In this framework, we identified sharing examples and sharing instructions as basic models and evaluated them on multiple tasks. Furthermore, we adapted the Secret Sharer metric to our framework, proposing a metric for measuring data leakage.

As next steps, we are looking for ways of improving the teaching process, for example by adding feedback loops and iteration. Furthermore, we want to investigate using social learning for modalities other than text.


Acknowledgements

We would like to acknowledge and thank Matt Sharifi, Sian Gooding, Lukas Zilka, and Blaise Aguera y Arcas, who are all co-authors on the paper. Furthermore, we would like to thank Victor Cărbune, Zachary Garrett, Tautvydas Misiunas, Sofia Neata and John Platt for their feedback, which greatly improved the paper. We’d also like to thank Tom Small for creating the animated figure.

Source: Google AI Blog


Croissant: a metadata format for ML-ready datasets

Posted by Omar Benjelloun, Software Engineer, Google Research, and Peter Mattson, Software Engineer, Google Core ML and President, MLCommons Association

Machine learning (ML) practitioners looking to reuse existing datasets to train an ML model often spend a lot of time understanding the data, making sense of its organization, or figuring out what subset to use as features. So much time, in fact, that progress in the field of ML is hampered by a fundamental obstacle: the wide variety of data representations.

ML datasets cover a broad range of content types, from text and structured data to images, audio, and video. Even within datasets that cover the same types of content, every dataset has a unique ad hoc arrangement of files and data formats. This challenge reduces productivity throughout the entire ML development process, from finding the data to training the model. It also impedes development of badly needed tooling for working with datasets.

There are general purpose metadata formats for datasets such as schema.org and DCAT. However, these formats were designed for data discovery rather than for the specific needs of ML data, such as the ability to extract and combine data from structured and unstructured sources, to include metadata that would enable responsible use of the data, or to describe ML usage characteristics such as defining training, test and validation sets.

Today, we're introducing Croissant, a new metadata format for ML-ready datasets. Croissant was developed collaboratively by a community from industry and academia, as part of the MLCommons effort. The Croissant format doesn't change how the actual data is represented (e.g., image or text file formats) — it provides a standard way to describe and organize it. Croissant builds upon schema.org, the de facto standard for publishing structured data on the Web, which is already used by over 40M datasets. Croissant augments it with comprehensive layers for ML relevant metadata, data resources, data organization, and default ML semantics.

In addition, we are announcing support from major tools and repositories: Today, three widely used collections of ML datasets — Kaggle, Hugging Face, and OpenML — will begin supporting the Croissant format for the datasets they host; the Dataset Search tool lets users search for Croissant datasets across the Web; and popular ML frameworks, including TensorFlow, PyTorch, and JAX, can load Croissant datasets easily using the TensorFlow Datasets (TFDS) package.


Croissant

This 1.0 release of Croissant includes a complete specification of the format, a set of example datasets, an open source Python library to validate, consume and generate Croissant metadata, and an open source visual editor to load, inspect and create Croissant dataset descriptions in an intuitive way.

Supporting Responsible AI (RAI) was a key goal of the Croissant effort from the start. We are also releasing the first version of the Croissant RAI vocabulary extension, which augments Croissant with key properties needed to describe important RAI use cases such as data life cycle management, data labeling, participatory data, ML safety and fairness evaluation, explainability, and compliance.


Why a shared format for ML data?

The majority of ML work is actually data work. The training data is the “code” that determines the behavior of a model. Datasets can vary from a collection of text used to train a large language model (LLM) to a collection of driving scenarios (annotated videos) used to train a car’s collision avoidance system. However, the steps to develop an ML model typically follow the same iterative data-centric process: (1) find or collect data, (2) clean and refine the data, (3) train the model on the data, (4) test the model on more data, (5) discover the model does not work, (6) analyze the data to find out why, (7) repeat until a workable model is achieved. Many steps are made harder by the lack of a common format. This “data development burden” is especially heavy for resource-limited research and early-stage entrepreneurial efforts.

The goal of a format like Croissant is to make this entire process easier. For instance, the metadata can be leveraged by search engines and dataset repositories to make it easier to find the right dataset. The data resources and organization information make it easier to develop tools for cleaning, refining, and analyzing data. This information and the default ML semantics make it possible for ML frameworks to use the data to train and test models with a minimum of code. Together, these improvements substantially reduce the data development burden.

Additionally, dataset authors care about the discoverability and ease of use of their datasets. Adopting Croissant improves the value of their datasets, while only requiring a minimal effort, thanks to the available creation tools and support from ML data platforms.


What can Croissant do today?

The Croissant ecosystem: Users can Search for Croissant datasets, download them from major repositories, and easily load them into their favorite ML frameworks. They can create, inspect and modify Croissant metadata using the Croissant editor.

Today, users can find Croissant datasets at:

With a Croissant dataset, it is possible to:

To publish a Croissant dataset, users can:

  • Use the Croissant editor UI (github) to generate a large portion of Croissant metadata automatically by analyzing the data the user provides, and to fill important metadata fields such as RAI properties.
  • Publish the Croissant information as part of their dataset Web page to make it discoverable and reusable.
  • Publish their data in one of the repositories that support Croissant, such as Kaggle, HuggingFace and OpenML, and automatically generate Croissant metadata.

Future direction

We are excited about Croissant's potential to help ML practitioners, but making this format truly useful requires the support of the community. We encourage dataset creators to consider providing Croissant metadata. We encourage platforms hosting datasets to provide Croissant files for download and embed Croissant metadata in dataset Web pages so that they can be made discoverable by dataset search engines. Tools that help users work with ML datasets, such as labeling or data analysis tools should also consider supporting Croissant datasets. Together, we can reduce the data development burden and enable a richer ecosystem of ML research and development.

We encourage the community to join us in contributing to the effort.


Acknowledgements

Croissant was developed by the Dataset Search, Kaggle and TensorFlow Datasets teams from Google, as part of an MLCommons community working group, which also includes contributors from these organizations: Bayer, cTuning Foundation, DANS-KNAW, Dotphoton, Harvard, Hugging Face, Kings College London, LIST, Meta, NASA, North Carolina State University, Open Data Institute, Open University of Catalonia, Sage Bionetworks, and TU Eindhoven.

Source: Google AI Blog


Google at APS 2024

Posted by Kate Weber and Shannon Leon, Google Research, Quantum AI Team

Today the 2024 March Meeting of the American Physical Society (APS) kicks off in Minneapolis, MN. A premier conference on topics ranging across physics and related fields, APS 2024 brings together researchers, students, and industry professionals to share their discoveries and build partnerships with the goal of realizing fundamental advances in physics-related sciences and technology.

This year, Google has a strong presence at APS with a booth hosted by the Google Quantum AI team, 50+ talks throughout the conference, and participation in conference organizing activities, special sessions and events. Attending APS 2024 in person? Come visit Google’s Quantum AI booth to learn more about the exciting work we’re doing to solve some of the field’s most interesting challenges.

You can learn more about the latest cutting edge work we are presenting at the conference along with our schedule of booth events below (Googlers listed in bold).


Organizing Committee

Session Chairs include: Aaron Szasz


Booth Activities

This schedule is subject to change. Please visit the Google Quantum AI booth for more information.

Crumble
Presenter: Matt McEwen
Tue, Mar 5 | 11:00 AM CST

Qualtran
Presenter: Tanuj Khattar
Tue, Mar 5 | 2:30 PM CST

Qualtran
Presenter: Tanuj Khattar
Thu, Mar 7 | 11:00 AM CST

$5M XPRIZE / Google Quantum AI competition to accelerate quantum applications Q&A
Presenter: Ryan Babbush
Thu, Mar 7 | 11:00 AM CST


Talks

Monday

Certifying highly-entangled states from few single-qubit measurements
Presenter: Hsin-Yuan Huang
Author: Hsin-Yuan Huang
Session A45: New Frontiers in Machine Learning Quantum Physics

Toward high-fidelity analog quantum simulation with superconducting qubits
Presenter: Trond Andersen
Authors: Trond I Andersen, Xiao Mi, Amir H Karamlou, Nikita Astrakhantsev, Andrey Klots, Julia Berndtsson, Andre Petukhov, Dmitry Abanin, Lev B Ioffe, Yu Chen, Vadim Smelyanskiy, Pedram Roushan
Session A51: Applications on Noisy Quantum Hardware I

Measuring circuit errors in context for surface code circuits
Presenter: Dripto M Debroy
Authors: Dripto M Debroy, Jonathan A Gross, Élie Genois, Zhang Jiang
Session B50: Characterizing Noise with QCVV Techniques

Quantum computation of stopping power for inertial fusion target design I: Physics overview and the limits of classical algorithms
Presenter: Andrew D. Baczewski
Authors: Nicholas C. Rubin, Dominic W. Berry, Alina Kononov, Fionn D. Malone, Tanuj Khattar, Alec White, Joonho Lee, Hartmut Neven, Ryan Babbush, Andrew D. Baczewski
Session B51: Heterogeneous Design for Quantum Applications
Link to Paper

Quantum computation of stopping power for inertial fusion target design II: Physics overview and the limits of classical algorithms
Presenter: Nicholas C. Rubin
Authors: Nicholas C. Rubin, Dominic W. Berry, Alina Kononov, Fionn D. Malone, Tanuj Khattar, Alec White, Joonho Lee, Hartmut Neven, Ryan Babbush, Andrew D. Baczewski
Session B51: Heterogeneous Design for Quantum Applications
Link to Paper

Calibrating Superconducting Qubits: From NISQ to Fault Tolerance
Presenter: Sabrina S Hong
Author: Sabrina S Hong
Session B56: From NISQ to Fault Tolerance

Measurement and feedforward induced entanglement negativity transition
Presenter: Ramis Movassagh
Authors: Alireza Seif, Yu-Xin Wang, Ramis Movassagh, Aashish A. Clerk
Session B31: Measurement Induced Criticality in Many-Body Systems
Link to Paper

Effective quantum volume, fidelity and computational cost of noisy quantum processing experiments
Presenter: Salvatore Mandra
Authors: Kostyantyn Kechedzhi, Sergei V Isakov, Salvatore Mandra, Benjamin Villalonga, X. Mi, Sergio Boixo, Vadim Smelyanskiy
Session B52: Quantum Algorithms and Complexity
Link to Paper

Accurate thermodynamic tables for solids using Machine Learning Interaction Potentials and Covariance of Atomic Positions
Presenter: Mgcini K Phuthi
Authors: Mgcini K Phuthi, Yang Huang, Michael Widom, Ekin D Cubuk, Venkat Viswanathan
Session D60: Machine Learning of Molecules and Materials: Chemical Space and Dynamics


Tuesday

IN-Situ Pulse Envelope Characterization Technique (INSPECT)
Presenter: Zhang Jiang
Authors: Zhang Jiang, Jonathan A Gross, Élie Genois
Session F50: Advanced Randomized Benchmarking and Gate Calibration

Characterizing two-qubit gates with dynamical decoupling
Presenter: Jonathan A Gross
Authors: Jonathan A Gross, Zhang Jiang, Élie Genois, Dripto M Debroy, Ze-Pei Cian*, Wojciech Mruczkiewicz
Session F50: Advanced Randomized Benchmarking and Gate Calibration

Statistical physics of regression with quadratic models
Presenter: Blake Bordelon
Authors: Blake Bordelon, Cengiz Pehlevan, Yasaman Bahri
Session EE01: V: Statistical and Nonlinear Physics II

Improved state preparation for first-quantized simulation of electronic structure
Presenter: William J Huggins
Authors: William J Huggins, Oskar Leimkuhler, Torin F Stetina, Birgitta Whaley
Session G51: Hamiltonian Simulation

Controlling large superconducting quantum processors
Presenter: Paul V. Klimov
Authors: Paul V. Klimov, Andreas Bengtsson, Chris Quintana, Alexandre Bourassa, Sabrina Hong, Andrew Dunsworth, Kevin J. Satzinger, William P. Livingston, Volodymyr Sivak, Murphy Y. Niu, Trond I. Andersen, Yaxing Zhang, Desmond Chik, Zijun Chen, Charles Neill, Catherine Erickson, Alejandro Grajales Dau, Anthony Megrant, Pedram Roushan, Alexander N. Korotkov, Julian Kelly, Vadim Smelyanskiy, Yu Chen, Hartmut Neven
Session G30: Commercial Applications of Quantum Computing)
Link to Paper

Gaussian boson sampling: Determining quantum advantage
Presenter: Peter D Drummond
Authors: Peter D Drummond, Alex Dellios, Ned Goodman, Margaret D Reid, Ben Villalonga
Session G50: Quantum Characterization, Verification, and Validation II

Attention to complexity III: learning the complexity of random quantum circuit states
Presenter: Hyejin Kim
Authors: Hyejin Kim, Yiqing Zhou, Yichen Xu, Chao Wan, Jin Zhou, Yuri D Lensky, Jesse Hoke, Pedram Roushan, Kilian Q Weinberger, Eun-Ah Kim
Session G50: Quantum Characterization, Verification, and Validation II

Balanced coupling in superconducting circuits
Presenter: Daniel T Sank
Authors: Daniel T Sank, Sergei V Isakov, Mostafa Khezri, Juan Atalaya
Session K48: Strongly Driven Superconducting Systems

Resource estimation of Fault Tolerant algorithms using Qᴜᴀʟᴛʀᴀɴ
Presenter: Tanuj Khattar
Author: Tanuj Khattar
Session K49: Algorithms and Implementations on Near-Term Quantum Computers


Wednesday

Discovering novel quantum dynamics with superconducting qubits
Presenter: Pedram Roushan
Author: Pedram Roushan
Session M24: Analog Quantum Simulations Across Platforms

Deciphering Tumor Heterogeneity in Triple-Negative Breast Cancer: The Crucial Role of Dynamic Cell-Cell and Cell-Matrix Interactions
Presenter: Susan Leggett
Authors: Susan Leggett, Ian Wong, Celeste Nelson, Molly Brennan, Mohak Patel, Christian Franck, Sophia Martinez, Joe Tien, Lena Gamboa, Thomas Valentin, Amanda Khoo, Evelyn K Williams
Session M27: Mechanics of Cells and Tissues II

Toward implementation of protected charge-parity qubits
Presenter: Abigail Shearrow
Authors: Abigail Shearrow, Matthew Snyder, Bradley G Cole, Kenneth R Dodge, Yebin Liu, Andrey Klots, Lev B Ioffe, Britton L Plourde, Robert McDermott
Session N48: Unconventional Superconducting Qubits

Electronic capacitance in tunnel junctions for protected charge-parity qubits
Presenter: Bradley G Cole
Authors: Bradley G Cole, Kenneth R Dodge, Yebin Liu, Abigail Shearrow, Matthew Snyder, Andrey Klots, Lev B Ioffe, Robert McDermott, B.L.T. Plourde
Session N48: Unconventional Superconducting Qubits

Overcoming leakage in quantum error correction
Presenter: Kevin C. Miao
Authors: Kevin C. Miao, Matt McEwen, Juan Atalaya, Dvir Kafri, Leonid P. Pryadko, Andreas Bengtsson, Alex Opremcak, Kevin J. Satzinger, Zijun Chen, Paul V. Klimov, Chris Quintana, Rajeev Acharya, Kyle Anderson, Markus Ansmann, Frank Arute, Kunal Arya, Abraham Asfaw, Joseph C. Bardin, Alexandre Bourassa, Jenna Bovaird, Leon Brill, Bob B. Buckley, David A. Buell, Tim Burger, Brian Burkett, Nicholas Bushnell, Juan Campero, Ben Chiaro, Roberto Collins, Paul Conner, Alexander L. Crook, Ben Curtin, Dripto M. Debroy, Sean Demura, Andrew Dunsworth, Catherine Erickson, Reza Fatemi, Vinicius S. Ferreira, Leslie Flores Burgos, Ebrahim Forati, Austin G. Fowler, Brooks Foxen, Gonzalo Garcia, William Giang, Craig Gidney, Marissa Giustina, Raja Gosula, Alejandro Grajales Dau, Jonathan A. Gross, Michael C. Hamilton, Sean D. Harrington, Paula Heu, Jeremy Hilton, Markus R. Hoffmann, Sabrina Hong, Trent Huang, Ashley Huff, Justin Iveland, Evan Jeffrey, Zhang Jiang, Cody Jones, Julian Kelly, Seon Kim, Fedor Kostritsa, John Mark Kreikebaum, David Landhuis, Pavel Laptev, Lily Laws, Kenny Lee, Brian J. Lester, Alexander T. Lill, Wayne Liu, Aditya Locharla, Erik Lucero, Steven Martin, Anthony Megrant, Xiao Mi, Shirin Montazeri, Alexis Morvan, Ofer Naaman, Matthew Neeley, Charles Neill, Ani Nersisyan, Michael Newman, Jiun How Ng, Anthony Nguyen, Murray Nguyen, Rebecca Potter, Charles Rocque, Pedram Roushan, Kannan Sankaragomathi, Christopher Schuster, Michael J. Shearn, Aaron Shorter, Noah Shutty, Vladimir Shvarts, Jindra Skruzny, W. Clarke Smith, George Sterling, Marco Szalay, Douglas Thor, Alfredo Torres, Theodore White, Bryan W. K. Woo, Z. Jamie Yao, Ping Yeh, Juhwan Yoo, Grayson Young, Adam Zalcman, Ningfeng Zhu, Nicholas Zobrist, Hartmut Neven, Vadim Smelyanskiy, Andre Petukhov, Alexander N. Korotkov, Daniel Sank, Yu Chen
Session N51: Quantum Error Correction Code Performance and Implementation I
Link to Paper

Modeling the performance of the surface code with non-uniform error distribution: Part 1
Presenter: Yuri D Lensky
Authors: Yuri D Lensky, Volodymyr Sivak, Kostyantyn Kechedzhi, Igor Aleiner
Session N51: Quantum Error Correction Code Performance and Implementation I

Modeling the performance of the surface code with non-uniform error distribution: Part 2
Presenter: Volodymyr Sivak
Authors: Volodymyr Sivak, Michael Newman, Cody Jones, Henry Schurkus, Dvir Kafri, Yuri D Lensky, Paul Klimov, Kostyantyn Kechedzhi, Vadim Smelyanskiy
Session N51: Quantum Error Correction Code Performance and Implementation I

Highly optimized tensor network contractions for the simulation of classically challenging quantum computations
Presenter: Benjamin Villalonga
Author: Benjamin Villalonga
Session Q51: Co-evolution of Quantum Classical Algorithms

Teaching modern quantum computing concepts using hands-on open-source software at all levels
Presenter: Abraham Asfaw
Author: Abraham Asfaw
Session Q61: Teaching Quantum Information at All Levels II


Thursday

New circuits and an open source decoder for the color code
Presenter: Craig Gidney
Authors: Craig Gidney, Cody Jones
Session S51: Quantum Error Correction Code Performance and Implementation II
Link to Paper

Performing Hartree-Fock many-body physics calculations with large language models
Presenter: Eun-Ah Kim
Authors: Eun-Ah Kim, Haining Pan, Nayantara Mudur, William Taranto, Subhashini Venugopalan, Yasaman Bahri, Michael P Brenner
Session S18: Data Science, AI and Machine Learning in Physics I

New methods for reducing resource overhead in the surface code
Presenter: Michael Newman
Authors: Craig M Gidney, Michael Newman, Peter Brooks, Cody Jones
Session S51: Quantum Error Correction Code Performance and Implementation II
Link to Paper

Challenges and opportunities for applying quantum computers to drug design
Presenter: Raffaele Santagati
Authors: Raffaele Santagati, Alan Aspuru-Guzik, Ryan Babbush, Matthias Degroote, Leticia Gonzalez, Elica Kyoseva, Nikolaj Moll, Markus Oppel, Robert M. Parrish, Nicholas C. Rubin, Michael Streif, Christofer S. Tautermann, Horst Weiss, Nathan Wiebe, Clemens Utschig-Utschig
Session S49: Advances in Quantum Algorithms for Near-Term Applications
Link to Paper

Dispatches from Google's hunt for super-quadratic quantum advantage in new applications
Presenter: Ryan Babbush
Author: Ryan Babbush
Session T45: Recent Advances in Quantum Algorithms

Qubit as a reflectometer
Presenter: Yaxing Zhang
Authors: Yaxing Zhang, Benjamin Chiaro
Session T48: Superconducting Fabrication, Packaging, & Validation

Random-matrix theory of measurement-induced phase transitions in nonlocal Floquet quantum circuits
Presenter: Aleksei Khindanov
Authors: Aleksei Khindanov, Lara Faoro, Lev Ioffe, Igor Aleiner
Session W14: Measurement-Induced Phase Transitions

Continuum limit of finite density many-body ground states with MERA
Presenter: Subhayan Sahu
Authors: Subhayan Sahu, Guifré Vidal
Session W58: Extreme-Scale Computational Science Discovery in Fluid Dynamics and Related Disciplines II

Dynamics of magnetization at infinite temperature in a Heisenberg spin chain
Presenter: Eliott Rosenberg
Authors: Eliott Rosenberg, Trond Andersen, Rhine Samajdar, Andre Petukhov, Jesse Hoke*, Dmitry Abanin, Andreas Bengtsson, Ilya Drozdov, Catherine Erickson, Paul Klimov, Xiao Mi, Alexis Morvan, Matthew Neeley, Charles Neill, Rajeev Acharya, Richard Allen, Kyle Anderson, Markus Ansmann, Frank Arute, Kunal Arya, Abraham Asfaw, Juan Atalaya, Joseph Bardin, A. Bilmes, Gina Bortoli, Alexandre Bourassa, Jenna Bovaird, Leon Brill, Michael Broughton, Bob B. Buckley, David Buell, Tim Burger, Brian Burkett, Nicholas Bushnell, Juan Campero, Hung-Shen Chang, Zijun Chen, Benjamin Chiaro, Desmond Chik, Josh Cogan, Roberto Collins, Paul Conner, William Courtney, Alexander Crook, Ben Curtin, Dripto Debroy, Alexander Del Toro Barba, Sean Demura, Agustin Di Paolo, Andrew Dunsworth, Clint Earle, E. Farhi, Reza Fatemi, Vinicius Ferreira, Leslie Flores, Ebrahim Forati, Austin Fowler, Brooks Foxen, Gonzalo Garcia, Élie Genois, William Giang, Craig Gidney, Dar Gilboa, Marissa Giustina, Raja Gosula, Alejandro Grajales Dau, Jonathan Gross, Steve Habegger, Michael Hamilton, Monica Hansen, Matthew Harrigan, Sean Harrington, Paula Heu, Gordon Hill, Markus Hoffmann, Sabrina Hong, Trent Huang, Ashley Huff, William Huggins, Lev Ioffe, Sergei Isakov, Justin Iveland, Evan Jeffrey, Zhang Jiang, Cody Jones, Pavol Juhas, D. Kafri, Tanuj Khattar, Mostafa Khezri, Mária Kieferová, Seon Kim, Alexei Kitaev, Andrey Klots, Alexander Korotkov, Fedor Kostritsa, John Mark Kreikebaum, David Landhuis, Pavel Laptev, Kim Ming Lau, Lily Laws, Joonho Lee, Kenneth Lee, Yuri Lensky, Brian Lester, Alexander Lill, Wayne Liu, William P. Livingston, A. Locharla, Salvatore Mandrà, Orion Martin, Steven Martin, Jarrod McClean, Matthew McEwen, Seneca Meeks, Kevin Miao, Amanda Mieszala, Shirin Montazeri, Ramis Movassagh, Wojciech Mruczkiewicz, Ani Nersisyan, Michael Newman, Jiun How Ng, Anthony Nguyen, Murray Nguyen, M. Niu, Thomas O'Brien, Seun Omonije, Alex Opremcak, Rebecca Potter, Leonid Pryadko, Chris Quintana, David Rhodes, Charles Rocque, N. Rubin, Negar Saei, Daniel Sank, Kannan Sankaragomathi, Kevin Satzinger, Henry Schurkus, Christopher Schuster, Michael Shearn, Aaron Shorter, Noah Shutty, Vladimir Shvarts, Volodymyr Sivak, Jindra Skruzny, Clarke Smith, Rolando Somma, George Sterling, Doug Strain, Marco Szalay, Douglas Thor, Alfredo Torres, Guifre Vidal, Benjamin Villalonga, Catherine Vollgraff Heidweiller, Theodore White, Bryan Woo, Cheng Xing, Jamie Yao, Ping Yeh, Juhwan Yoo, Grayson Young, Adam Zalcman, Yaxing Zhang, Ningfeng Zhu, Nicholas Zobrist, Hartmut Neven, Ryan Babbush, Dave Bacon, Sergio Boixo, Jeremy Hilton, Erik Lucero, Anthony Megrant, Julian Kelly, Yu Chen, Vadim Smelyanskiy, Vedika Khemani, Sarang Gopalakrishnan, Tomaž Prosen, Pedram Roushan
Session W50: Quantum Simulation of Many-Body Physics
Link to Paper

The fast multipole method on a quantum computer
Presenter: Kianna Wan
Authors: Kianna Wan, Dominic W Berry, Ryan Babbush
Session W50: Quantum Simulation of Many-Body Physics


Friday

The quantum computing industry and protecting national security: what tools will work?
Presenter: Kate Weber
Author: Kate Weber
Session Y43: Industry, Innovation, and National Security: Finding the Right Balance

Novel charging effects in the fluxonium qubit
Presenter: Agustin Di Paolo
Authors: Agustin Di Paolo, Kyle Serniak, Andrew J Kerman, William D Oliver
Session Y46: Fluxonium-Based Superconducting Quibits

Microwave Engineering of Parametric Interactions in Superconducting Circuits
Presenter: Ofer Naaman
Author: Ofer Naaman
Session Z46: Broadband Parametric Amplifiers and Circulators

Linear spin wave theory of large magnetic unit cells using the Kernel Polynomial Method
Presenter: Harry Lane
Authors: Harry Lane, Hao Zhang, David A Dahlbom, Sam Quinn, Rolando D Somma, Martin P Mourigal, Cristian D Batista, Kipton Barros
Session Z62: Cooperative Phenomena, Theory

*Work done while at Google

Source: Google AI Blog


VideoPrism: A foundational visual encoder for video understanding

Posted by Long Zhao, Senior Research Scientist, and Ting Liu, Senior Staff Software Engineer, Google Research

An astounding number of videos are available on the Web, covering a variety of content from everyday moments people share to historical moments to scientific observations, each of which contains a unique record of the world. The right tools could help researchers analyze these videos, transforming how we understand the world around us.

Videos offer dynamic visual content far more rich than static images, capturing movement, changes, and dynamic relationships between entities. Analyzing this complexity, along with the immense diversity of publicly available video data, demands models that go beyond traditional image understanding. Consequently, many of the approaches that best perform on video understanding still rely on specialized models tailor-made for particular tasks. Recently, there has been exciting progress in this area using video foundation models (ViFMs), such as VideoCLIP, InternVideo, VideoCoCa, and UMT). However, building a ViFM that handles the sheer diversity of video data remains a challenge.

With the goal of building a single model for general-purpose video understanding, we introduced “VideoPrism: A Foundational Visual Encoder for Video Understanding”. VideoPrism is a ViFM designed to handle a wide spectrum of video understanding tasks, including classification, localization, retrieval, captioning, and question answering (QA). We propose innovations in both the pre-training data as well as the modeling strategy. We pre-train VideoPrism on a massive and diverse dataset: 36 million high-quality video-text pairs and 582 million video clips with noisy or machine-generated parallel text. Our pre-training approach is designed for this hybrid data, to learn both from video-text pairs and the videos themselves. VideoPrism is incredibly easy to adapt to new video understanding challenges, and achieves state-of-the-art performance using a single frozen model.

VideoPrism is a general-purpose video encoder that enables state-of-the-art results over a wide spectrum of video understanding tasks, including classification, localization, retrieval, captioning, and question answering, by producing video representations from a single frozen model.

Pre-training data

A powerful ViFM needs a very large collection of videos on which to train — similar to other foundation models (FMs), such as those for large language models (LLMs). Ideally, we would want the pre-training data to be a representative sample of all the videos in the world. While naturally most of these videos do not have perfect captions or descriptions, even imperfect text can provide useful information about the semantic content of the video.

To give our model the best possible starting point, we put together a massive pre-training corpus consisting of several public and private datasets, including YT-Temporal-180M, InternVid, VideoCC, WTS-70M, etc. This includes 36 million carefully selected videos with high-quality captions, along with an additional 582 million clips with varying levels of noisy text (like auto-generated transcripts). To our knowledge, this is the largest and most diverse video training corpus of its kind.

Statistics on the video-text pre-training data. The large variations of the CLIP similarity scores (the higher, the better) demonstrate the diverse caption quality of our pre-training data, which is a byproduct of the various ways used to harvest the text.

Two-stage training

The VideoPrism model architecture stems from the standard vision transformer (ViT) with a factorized design that sequentially encodes spatial and temporal information following ViViT. Our training approach leverages both the high-quality video-text data and the video data with noisy text mentioned above. To start, we use contrastive learning (an approach that minimizes the distance between positive video-text pairs while maximizing the distance between negative video-text pairs) to teach our model to match videos with their own text descriptions, including imperfect ones. This builds a foundation for matching semantic language content to visual content.

After video-text contrastive training, we leverage the collection of videos without text descriptions. Here, we build on the masked video modeling framework to predict masked patches in a video, with a few improvements. We train the model to predict both the video-level global embedding and token-wise embeddings from the first-stage model to effectively leverage the knowledge acquired in that stage. We then randomly shuffle the predicted tokens to prevent the model from learning shortcuts.

What is unique about VideoPrism’s setup is that we use two complementary pre-training signals: text descriptions and the visual content within a video. Text descriptions often focus on what things look like, while the video content provides information about movement and visual dynamics. This enables VideoPrism to excel in tasks that demand an understanding of both appearance and motion.


Results

We conducted extensive evaluation on VideoPrism across four broad categories of video understanding tasks, including video classification and localization, video-text retrieval, video captioning, question answering, and scientific video understanding. VideoPrism achieves state-of-the-art performance on 30 out of 33 video understanding benchmarks — all with minimal adaptation of a single, frozen model.

VideoPrism compared to the previous best-performing FMs.

Classification and localization

We evaluate VideoPrism on an existing large-scale video understanding benchmark (VideoGLUE) covering classification and localization tasks. We found that (1) VideoPrism outperforms all of the other state-of-the-art FMs, and (2) no other single model consistently came in second place. This tells us that VideoPrism has learned to effectively pack a variety of video signals into one encoder — from semantics at different granularities to appearance and motion cues — and it works well across a variety of video sources.

VideoPrism outperforms state-of-the-art approaches (including CLIP, VATT, InternVideo, and UMT) on the video understanding benchmark. In this plot, we show the absolute score differences compared with the previous best model to highlight the relative improvements of VideoPrism. On Charades, ActivityNet, AVA, and AVA-K, we use mean average precision (mAP) as the evaluation metric. On the other datasets, we report top-1 accuracy.

Combining with LLMs

We further explore combining VideoPrism with LLMs to unlock its ability to handle various video-language tasks. In particular, when paired with a text encoder (following LiT) or a language decoder (such as PaLM-2), VideoPrism can be utilized for video-text retrieval, video captioning, and video QA tasks. We compare the combined models on a broad and challenging set of vision-language benchmarks. VideoPrism sets the new state of the art on most benchmarks. From the visual results, we find that VideoPrism is capable of understanding complex motions and appearances in videos (e.g., the model can recognize the different colors of spinning objects on the window in the visual examples below). These results demonstrate that VideoPrism is strongly compatible with language models.

VideoPrism achieves competitive results compared with state-of-the-art approaches (including VideoCoCa, UMT and Flamingo) on multiple video-text retrieval (top) and video captioning and video QA (bottom) benchmarks. We also show the absolute score differences compared with the previous best model to highlight the relative improvements of VideoPrism. We report the Recall@1 on MASRVTT, VATEX, and ActivityNet, CIDEr score on MSRVTT-Cap, VATEX-Cap, and YouCook2, top-1 accuracy on MSRVTT-QA and MSVD-QA, and WUPS index on NExT-QA.

We show qualitative results using VideoPrism with a text encoder for video-text retrieval (first row) and adapted to a language decoder for video QA (second and third row). For video-text retrieval examples, the blue bars indicate the embedding similarities between the videos and the text queries.

Scientific applications

Finally, we tested VideoPrism on datasets used by scientists across domains, including fields such as ethology, behavioral neuroscience, and ecology. These datasets typically require domain expertise to annotate, for which we leverage existing scientific datasets open-sourced by the community including Fly vs. Fly, CalMS21, ChimpACT, and KABR. VideoPrism not only performs exceptionally well, but actually surpasses models designed specifically for those tasks. This suggests tools like VideoPrism have the potential to transform how scientists analyze video data across different fields.

VideoPrism outperforms the domain experts on various scientific benchmarks. We show the absolute score differences to highlight the relative improvements of VideoPrism. We report mean average precision (mAP) for all datasets, except for KABR which uses class-averaged top-1 accuracy.

Conclusion

With VideoPrism, we introduce a powerful and versatile video encoder that sets a new standard for general-purpose video understanding. Our emphasis on both building a massive and varied pre-training dataset and innovative modeling techniques has been validated through our extensive evaluations. Not only does VideoPrism consistently outperform strong baselines, but its unique ability to generalize positions it well for tackling an array of real-world applications. Because of its potential broad use, we are committed to continuing further responsible research in this space, guided by our AI Principles. We hope VideoPrism paves the way for future breakthroughs at the intersection of AI and video analysis, helping to realize the potential of ViFMs across domains such as scientific discovery, education, and healthcare.


Acknowledgements

This blog post is made on behalf of all the VideoPrism authors: Long Zhao, Nitesh B. Gundavarapu, Liangzhe Yuan, Hao Zhou, Shen Yan, Jennifer J. Sun, Luke Friedman, Rui Qian, Tobias Weyand, Yue Zhao, Rachel Hornung, Florian Schroff, Ming-Hsuan Yang, David A. Ross, Huisheng Wang, Hartwig Adam, Mikhail Sirotenko, Ting Liu, and Boqing Gong. We sincerely thank David Hendon for their product management efforts, and Alex Siegman, Ramya Ganeshan, and Victor Gomes for their program and resource management efforts. We also thank Hassan Akbari, Sherry Ben, Yoni Ben-Meshulam, Chun-Te Chu, Sam Clearwater, Yin Cui, Ilya Figotin, Anja Hauth, Sergey Ioffe, Xuhui Jia, Yeqing Li, Lu Jiang, Zu Kim, Dan Kondratyuk, Bill Mark, Arsha Nagrani, Caroline Pantofaru, Sushant Prakash, Cordelia Schmid, Bryan Seybold, Mojtaba Seyedhosseini, Amanda Sadler, Rif A. Saurous, Rachel Stigler, Paul Voigtlaender, Pingmei Xu, Chaochao Yan, Xuan Yang, and Yukun Zhu for the discussions, support, and feedback that greatly contributed to this work. We are grateful to Jay Yagnik, Rahul Sukthankar, and Tomas Izo for their enthusiastic support for this project. Lastly, we thank Tom Small, Jennifer J. Sun, Hao Zhou, Nitesh B. Gundavarapu, Luke Friedman, and Mikhail Sirotenko for the tremendous help with making this blog post.

Source: Google AI Blog


Advances in private training for production on-device language models

Posted by Zheng Xu, Research Scientist, and Yanxiang Zhang, Software Engineer, Google

Language models (LMs) trained to predict the next word given input text are the key technology for many applications [1, 2]. In Gboard, LMs are used to improve users’ typing experience by supporting features like next word prediction (NWP), Smart Compose, smart completion and suggestion, slide to type, and proofread. Deploying models on users’ devices rather than enterprise servers has advantages like lower latency and better privacy for model usage. While training on-device models directly from user data effectively improves the utility performance for applications such as NWP and smart text selection, protecting the privacy of user data for model training is important.

Gboard features powered by on-device language models.

In this blog we discuss how years of research advances now power the private training of Gboard LMs, since the proof-of-concept development of federated learning (FL) in 2017 and formal differential privacy (DP) guarantees in 2022. FL enables mobile phones to collaboratively learn a model while keeping all the training data on device, and DP provides a quantifiable measure of data anonymization. Formally, DP is often characterized by (ε, δ) with smaller values representing stronger guarantees. Machine learning (ML) models are considered to have reasonable DP guarantees for ε=10 and strong DP guarantees for ε=1 when δ is small.

As of today, all NWP neural network LMs in Gboard are trained with FL with formal DP guarantees, and all future launches of Gboard LMs trained on user data require DP. These 30+ Gboard on-device LMs are launched in 7+ languages and 15+ countries, and satisfy (ɛ, δ)-DP guarantees of small δ of 10-10 and ɛ between 0.994 and 13.69. To the best of our knowledge, this is the largest known deployment of user-level DP in production at Google or anywhere, and the first time a strong DP guarantee of ɛ < 1 is announced for models trained directly on user data.


Privacy principles and practices in Gboard

In “Private Federated Learning in Gboard”, we discussed how different privacy principles are currently reflected in production models, including:

  • Transparency and user control: We provide disclosure of what data is used, what purpose it is used for, how it is processed in various channels, and how Gboard users can easily configure the data usage in learning models.
  • Data minimization: FL immediately aggregates only focused updates that improve a specific model. Secure aggregation (SecAgg) is an encryption method to further guarantee that only aggregated results of the ephemeral updates can be accessed.
  • Data anonymization: DP is applied by the server to prevent models from memorizing the unique information in individual user’s training data.
  • Auditability and verifiability: We have made public the key algorithmic approaches and privacy accounting in open-sourced code (TFF aggregator, TFP DPQuery, DP accounting, and FL system).

A brief history

In recent years, FL has become the default method for training Gboard on-device LMs from user data. In 2020, a DP mechanism that clips and adds noise to model updates was used to prevent memorization for training the Spanish LM in Spain, which satisfies finite DP guarantees (Tier 3 described in “How to DP-fy ML“ guide). In 2022, with the help of the DP-Follow-The-Regularized-Leader (DP-FTRL) algorithm, the Spanish LM became the first production neural network trained directly on user data announced with a formal DP guarantee of (ε=8.9, δ=10-10)-DP (equivalent to the reported ρ=0.81 zero-Concentrated-Differential-Privacy), and therefore satisfies reasonable privacy guarantees (Tier 2).


Differential privacy by default in federated learning

In “Federated Learning of Gboard Language Models with Differential Privacy”, we announced that all the NWP neural network LMs in Gboard have DP guarantees, and all future launches of Gboard LMs trained on user data require DP guarantees. DP is enabled in FL by applying the following practices:

  • Pre-train the model with the multilingual C4 dataset.
  • Via simulation experiments on public datasets, find a large DP-noise-to-signal ratio that allows for high utility. Increasing the number of clients contributing to one round of model update improves privacy while keeping the noise ratio fixed for good utility, up to the point the DP target is met, or the maximum allowed by the system and the size of the population.
  • Configure the parameter to restrict the frequency each client can contribute (e.g., once every few days) based on computation budget and estimated population in the FL system.
  • Run DP-FTRL training with limits on the magnitude of per-device updates chosen either via adaptive clipping, or fixed based on experience.

SecAgg can be additionally applied by adopting the advances in improving computation and communication for scales and sensitivity.

Federated learning with differential privacy and (SecAgg).

Reporting DP guarantees

The DP guarantees of launched Gboard NWP LMs are visualized in the barplot below. The x-axis shows LMs labeled by language-locale and trained on corresponding populations; the y-axis shows the ε value when δ is fixed to a small value of 10-10 for (ε, δ)-DP (lower is better). The utility of these models are either significantly better than previous non-neural models in production, or comparable with previous LMs without DP, measured based on user-interactions metrics during A/B testing. For example, by applying the best practices, the DP guarantee of the Spanish model in Spain is improved from ε=8.9 to ε=5.37. SecAgg is additionally used for training the Spanish model in Spain and English model in the US. More details of the DP guarantees are reported in the appendix following the guidelines outlined in “How to DP-fy ML”.


Towards stronger DP guarantees

The ε~10 DP guarantees of many launched LMs are already considered reasonable for ML models in practice, while the journey of DP FL in Gboard continues for improving user typing experience while protecting data privacy. We are excited to announce that, for the first time, production LMs of Portuguese in Brazil and Spanish in Latin America are trained and launched with a DP guarantee of ε ≤ 1, which satisfies Tier 1 strong privacy guarantees. Specifically, the (ε=0.994, δ=10-10)-DP guarantee is achieved by running the advanced Matrix Factorization DP-FTRL (MF-DP-FTRL) algorithm, with 12,000+ devices participating in every training round of server model update larger than the common setting of 6500+ devices, and a carefully configured policy to restrict each client to at most participate twice in the total 2000 rounds of training in 14 days in the large Portuguese user population of Brazil. Using a similar setting, the es-US Spanish LM was trained in a large population combining multiple countries in Latin America to achieve (ε=0.994, δ=10-10)-DP. The ε ≤ 1 es-US model significantly improved the utility in many countries, and launched in Colombia, Ecuador, Guatemala, Mexico, and Venezuela. For the smaller population in Spain, the DP guarantee of es-ES LM is improved from ε=5.37 to ε=3.42 by only replacing DP-FTRL with MF-DP-FTRL without increasing the number of devices participating every round. More technical details are disclosed in the colab for privacy accounting.

DP guarantees for Gboard NWP LMs (the purple bar represents the first es-ES launch of ε=8.9; cyan bars represent privacy improvements for models trained with MF-DP-FTRL; tiers are from “How to DP-fy ML“ guide; en-US* and es-ES* are additionally trained with SecAgg).

Discussion and next steps

Our experience suggests that DP can be achieved in practice through system algorithm co-design on client participation, and that both privacy and utility can be strong when populations are large and a large number of devices' contributions are aggregated. Privacy-utility-computation trade-offs can be improved by using public data, the new MF-DP-FTRL algorithm, and tightening accounting. With these techniques, a strong DP guarantee of ε ≤ 1 is possible but still challenging. Active research on empirical privacy auditing [1, 2] suggests that DP models are potentially more private than the worst-case DP guarantees imply. While we keep pushing the frontier of algorithms, which dimension of privacy-utility-computation should be prioritized?

We are actively working on all privacy aspects of ML, including extending DP-FTRL to distributed DP and improving auditability and verifiability. Trusted Execution Environment opens the opportunity for substantially increasing the model size with verifiable privacy. The recent breakthrough in large LMs (LLMs) motivates us to rethink the usage of public information in private training and more future interactions between LLMs, on-device LMs, and Gboard production.


Acknowledgments

The authors would like to thank Peter Kairouz, Brendan McMahan, and Daniel Ramage for their early feedback on the blog post itself, Shaofeng Li and Tom Small for helping with the animated figures, and the teams at Google that helped with algorithm design, infrastructure implementation, and production maintenance. The collaborators below directly contribute to the presented results:

Research and algorithm development: Galen Andrew, Stanislav Chiknavaryan, Christopher A. Choquette-Choo, Arun Ganesh, Peter Kairouz, Ryan McKenna, H. Brendan McMahan, Jesse Rosenstock, Timon Van Overveldt, Keith Rush, Shuang Song, Thomas Steinke, Abhradeep Guha Thakurta, Om Thakkar, and Yuanbo Zhang.

Infrastructure, production and leadership support: Mingqing Chen, Stefan Dierauf, Billy Dou, Hubert Eichner, Zachary Garrett, Jeremy Gillula, Jianpeng Hou, Hui Li, Xu Liu, Wenzhi Mao, Brett McLarnon, Mengchen Pei, Daniel Ramage, Swaroop Ramaswamy, Haicheng Sun, Andreas Terzis, Yun Wang, Shanshan Wu, Yu Xiao, and Shumin Zhai.

Source: Google AI Blog


Learning the importance of training data under concept drift

Posted by Nishant Jain, Pre-doctoral Researcher, and Pradeep Shenoy, Research Scientist, Google Research

The constantly changing nature of the world around us poses a significant challenge for the development of AI models. Often, models are trained on longitudinal data with the hope that the training data used will accurately represent inputs the model may receive in the future. More generally, the default assumption that all training data are equally relevant often breaks in practice. For example, the figure below shows images from the CLEAR nonstationary learning benchmark, and it illustrates how visual features of objects evolve significantly over a 10 year span (a phenomenon we refer to as slow concept drift), posing a challenge for object categorization models.

Sample images from the CLEAR benchmark. (Adapted from Lin et al.)

Alternative approaches, such as online and continual learning, repeatedly update a model with small amounts of recent data in order to keep it current. This implicitly prioritizes recent data, as the learnings from past data are gradually erased by subsequent updates. However in the real world, different kinds of information lose relevance at different rates, so there are two key issues: 1) By design they focus exclusively on the most recent data and lose any signal from older data that is erased. 2) Contributions from data instances decay uniformly over time irrespective of the contents of the data.

In our recent work, “Instance-Conditional Timescales of Decay for Non-Stationary Learning”, we propose to assign each instance an importance score during training in order to maximize model performance on future data. To accomplish this, we employ an auxiliary model that produces these scores using the training instance as well as its age. This model is jointly learned with the primary model. We address both the above challenges and achieve significant gains over other robust learning methods on a range of benchmark datasets for nonstationary learning. For instance, on a recent large-scale benchmark for nonstationary learning (~39M photos over a 10 year period), we show up to 15% relative accuracy gains through learned reweighting of training data.


The challenge of concept drift for supervised learning

To gain quantitative insight into slow concept drift, we built classifiers on a recent photo categorization task, comprising roughly 39M photographs sourced from social media websites over a 10 year period. We compared offline training, which iterated over all the training data multiple times in random order, and continual training, which iterated multiple times over each month of data in sequential (temporal) order. We measured model accuracy both during the training period and during a subsequent period where both models were frozen, i.e., not updated further on new data (shown below). At the end of the training period (left panel, x-axis = 0), both approaches have seen the same amount of data, but show a large performance gap. This is due to catastrophic forgetting, a problem in continual learning where a model’s knowledge of data from early on in the training sequence is diminished in an uncontrolled manner. On the other hand, forgetting has its advantages — over the test period (shown on the right), the continual trained model degrades much less rapidly than the offline model because it is less dependent on older data. The decay of both models’ accuracy in the test period is confirmation that the data is indeed evolving over time, and both models become increasingly less relevant.

Comparing offline and continually trained models on the photo classification task.


Time-sensitive reweighting of training data

We design a method combining the benefits of offline learning (the flexibility of effectively reusing all available data) and continual learning (the ability to downplay older data) to address slow concept drift. We build upon offline learning, then add careful control over the influence of past data and an optimization objective, both designed to reduce model decay in the future.

Suppose we wish to train a model, M, given some training data collected over time. We propose to also train a helper model that assigns a weight to each point based on its contents and age. This weight scales the contribution from that data point in the training objective for M. The objective of the weights is to improve the performance of M on future data.

In our work, we describe how the helper model can be meta-learned, i.e., learned alongside M in a manner that helps the learning of the model M itself. A key design choice of the helper model is that we separated out instance- and age-related contributions in a factored manner. Specifically, we set the weight by combining contributions from multiple different fixed timescales of decay, and learn an approximate “assignment” of a given instance to its most suited timescales. We find in our experiments that this form of the helper model outperforms many other alternatives we considered, ranging from unconstrained joint functions to a single timescale of decay (exponential or linear), due to its combination of simplicity and expressivity. Full details may be found in the paper.


Instance weight scoring

The top figure below shows that our learned helper model indeed up-weights more modern-looking objects in the CLEAR object recognition challenge; older-looking objects are correspondingly down-weighted. On closer examination (bottom figure below, gradient-based feature importance assessment), we see that the helper model focuses on the primary object within the image, as opposed to, e.g., background features that may spuriously be correlated with instance age.

Sample images from the CLEAR benchmark (camera & computer categories) assigned the highest and lowest weights respectively by our helper model.

Feature importance analysis of our helper model on sample images from the CLEAR benchmark.


Results


Gains on large-scale data

We first study the large-scale photo categorization task (PCAT) on the YFCC100M dataset discussed earlier, using the first five years of data for training and the next five years as test data. Our method (shown in red below) improves substantially over the no-reweighting baseline (black) as well as many other robust learning techniques. Interestingly, our method deliberately trades off accuracy on the distant past (training data unlikely to reoccur in the future) in exchange for marked improvements in the test period. Also, as desired, our method degrades less than other baselines in the test period.

Comparison of our method and relevant baselines on the PCAT dataset.


Broad applicability

We validated our findings on a wide range of nonstationary learning challenge datasets sourced from the academic literature (see 1, 2, 3, 4 for details) that spans data sources and modalities (photos, satellite images, social media text, medical records, sensor readings, tabular data) and sizes (ranging from 10k to 39M instances). We report significant gains in the test period when compared to the nearest published benchmark method for each dataset (shown below). Note that the previous best-known method may be different for each dataset. These results showcase the broad applicability of our approach.

Performance gain of our method on a variety of tasks studying natural concept drift. Our reported gains are over the previous best-known method for each dataset.


Extensions to continual learning

Finally, we consider an interesting extension of our work. The work above described how offline learning can be extended to handle concept drift using ideas inspired by continual learning. However, sometimes offline learning is infeasible — for example, if the amount of training data available is too large to maintain or process. We adapted our approach to continual learning in a straightforward manner by applying temporal reweighting within the context of each bucket of data being used to sequentially update the model. This proposal still retains some limitations of continual learning, e.g., model updates are performed only on most-recent data, and all optimization decisions (including our reweighting) are only made over that data. Nevertheless, our approach consistently beats regular continual learning as well as a wide range of other continual learning algorithms on the photo categorization benchmark (see below). Since our approach is complementary to the ideas in many baselines compared here, we anticipate even larger gains when combined with them.

Results of our method adapted to continual learning, compared to the latest baselines.


Conclusion

We addressed the challenge of data drift in learning by combining the strengths of previous approaches — offline learning with its effective reuse of data, and continual learning with its emphasis on more recent data. We hope that our work helps improve model robustness to concept drift in practice, and generates increased interest and new ideas in addressing the ubiquitous problem of slow concept drift.


Acknowledgements

We thank Mike Mozer for many interesting discussions in the early phase of this work, as well as very helpful advice and feedback during its development.

Source: Google AI Blog