Publications (56)
The Flan Collection: Designing Data and Methods for Effective Instruction Tuning
Shayne Longpre, Le Hou, Tu Vu +8
We study the design decisions of publicly available instruction tuning methods, and break down the development of Flan 2022 (Chung et al., 2022). Through careful ablation studies o…
A Way out of the Odyssey: Analyzing and Combining Recent Insights for LSTMs
Shayne Longpre, Sabeek Pradhan, Caiming Xiong +1
LSTMs have become a basic building block for many deep NLP models. In recent years, many improvements and variations have been proposed for deep sequence models in general, and LST…
BLOOM: A 176B-Parameter Open-Access Multilingual Language Model
BigScience Workshop, :, Teven Le Scao +391
Large language models (LLMs) have been shown to be able to perform new tasks based on a few demonstrations or natural language instructions. While these capabilities have led to wi…
ThoughtTrace: Understanding User Thoughts in Real-World LLM Interactions
Chuanyang Jin, Binze Li, Haopeng Xie +6
Conversational AI has now reached billions of users, yet existing datasets capture only what people say, not what they think. We introduce ThoughtTrace, the first large-scale datas…
A Systematic Review of NeurIPS Dataset Management Practices
Yiwei Wu, Leah Ajmani, Shayne Longpre +1
As new machine learning methods demand larger training datasets, researchers and developers face significant challenges in dataset management. Although ethics reviews, documentatio…
A Safe Harbor for AI Evaluation and Red Teaming
Shayne Longpre, Sayash Kapoor, Kevin Klyman +20
Independent evaluation and red teaming are critical for identifying the risks posed by generative AI systems. However, the terms of service and enforcement strategies used by promi…
MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question Answering
Shayne Longpre, Yi Lu, Joachim Daiber
Progress in cross-lingual modeling depends on challenging, realistic, and diverse evaluation sets. We introduce Multilingual Knowledge Questions and Answers (MKQA), an open-domain…
A Comparison of Question Rewriting Methods for Conversational Passage Retrieval
Svitlana Vakulenko, Nikos Voskarides, Zhucheng Tu +1
Conversational passage retrieval relies on question rewriting to modify the original question so that it no longer depends on the conversation history. Several methods for question…
FlexOlmo: Open Language Models for Flexible Data Use
Weijia Shi, Akshita Bhagia, Kevin Farhat +20
We introduce FlexOlmo, a new class of language models (LMs) that supports (1) distributed training without data sharing, where different model parameters are independently trained…
Establishing Best Practices for Building Rigorous Agentic Benchmarks
Yuxuan Zhu, Tengjun Jin, Yada Pruksachatkun +22
Benchmarks are essential for quantitatively tracking progress in AI. As AI agents become increasingly capable, researchers and practitioners have introduced agentic benchmarks to e…
To Err is AI : A Case Study Informing LLM Flaw Reporting Practices
Sean McGregor, Allyson Ettinger, Nick Judd +10
In August of 2024, 495 hackers generated evaluations in an open-ended bug bounty targeting the Open Language Model (OLMo) from The Allen Institute for AI. A vendor panel staffed by…
MIA 2022 Shared Task: Evaluating Cross-lingual Open-Retrieval Question Answering for 16 Diverse Languages
Akari Asai, Shayne Longpre, Jungo Kasai +6
We present the results of the Workshop on Multilingual Information Access (MIA) 2022 Shared Task, evaluating cross-lingual open-retrieval question answering (QA) systems in 16 typo…
Towards Best Practices for Open Datasets for LLM Training
Stefan Baack, Stella Biderman, Kasia Odrozek +36
Many AI companies are training their large language models (LLMs) on data without the permission of the copyright owners. The permissibility of doing so varies by jurisdiction: in…
Mixture-of-Experts Meets Instruction Tuning:A Winning Combination for Large Language Models
Sheng Shen, Le Hou, Yanqi Zhou +17
Sparse Mixture-of-Experts (MoE) is a neural architecture design that can be utilized to add learnable parameters to Large Language Models (LLMs) without increasing inference cost.…
AI-Powered Autonomous Weapons Risk Geopolitical Instability and Threaten AI Research
Riley Simmons-Edler, Ryan Badman, Shayne Longpre +1
The recent embrace of machine learning (ML) in the development of autonomous weapons systems (AWS) creates serious risks to geopolitical stability and the free exchange of ideas in…
Consent in Crisis: The Rapid Decline of the AI Data Commons
Shayne Longpre, Robert Mahari, Ariel Lee +46
General-purpose artificial intelligence (AI) systems are built on massive swathes of public web data, assembled into corpora such as C4, RefinedWeb, and Dolma. To our knowledge, we…
Leveraging Query Resolution and Reading Comprehension for Conversational Passage Retrieval
Svitlana Vakulenko, Nikos Voskarides, Zhucheng Tu +1
This paper describes the participation of UvA.ILPS group at the TREC CAsT 2020 track. Our passage retrieval pipeline consists of (i) an initial retrieval module that uses BM25, and…
In-House Evaluation Is Not Enough: Towards Robust Third-Party Flaw Disclosure for General-Purpose AI
Shayne Longpre, Kevin Klyman, Ruth E. Appel +31
The widespread deployment of general-purpose AI (GPAI) systems introduces significant new risks. Yet the infrastructure, practices, and norms for reporting flaws in GPAI systems re…
ATLAS: Adaptive Transfer Scaling Laws for Multilingual Pretraining, Finetuning, and Decoding the Curse of Multilinguality
Shayne Longpre, Sneha Kudugunta, Niklas Muennighoff +6
Scaling laws research has focused overwhelmingly on English -- yet the most prominent AI models explicitly serve billions of international users. In this work, we undertake the lar…
International AI Safety Report
Yoshua Bengio, Sören Mindermann, Daniel Privitera +93
The first International AI Safety Report comprehensively synthesizes the current evidence on the capabilities, risks, and safety of advanced AI systems. The report was mandated by…
The 2025 Foundation Model Transparency Index
Alexander Wan, Kevin Klyman, Sayash Kapoor +5
Foundation model developers are among the world's most important companies. As these companies become increasingly consequential, how do their transparency practices evolve? The 20…
The BiGGen Bench: A Principled Benchmark for Fine-grained Evaluation of Language Models with Language Models
Seungone Kim, Juyoung Suk, Ji Yong Cho +29
As language models (LMs) become capable of handling a wide range of tasks, their evaluation is becoming as challenging as their development. Most generation benchmarks currently as…
Bridging the Data Provenance Gap Across Text, Speech and Video
Shayne Longpre, Nikhil Singh, Manuel Cherep +40
Progress in AI is driven largely by the scale and quality of training data. Despite this, there is a deficit of empirical analysis examining the attributes of well-established data…
Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models
Seungone Kim, Juyoung Suk, Shayne Longpre +7
Proprietary LMs such as GPT-4 are often employed to assess the quality of responses from various LMs. However, concerns including transparency, controllability, and affordability s…
Economies of Open Intelligence: Tracing Power & Participation in the Model Ecosystem
Shayne Longpre, Christopher Akiki, Campbell Lund +6
Since 2019, the Hugging Face Model Hub has been the primary global platform for sharing open weight AI models. By releasing a dataset of the complete history of weekly model downlo…
OctoPack: Instruction Tuning Code Large Language Models
Niklas Muennighoff, Qian Liu, Armel Zebaze +7
Finetuning large language models (LLMs) on instructions leads to vast performance improvements on natural language tasks. We apply instruction tuning using code, leveraging the nat…
Entity-Based Knowledge Conflicts in Question Answering
Shayne Longpre, Kartik Perisetla, Anthony Chen +3
Knowledge-dependent tasks typically use two sources of knowledge: parametric, learned at training time, and contextual, given as a passage at inference time. To understand how mode…
How Effective is Task-Agnostic Data Augmentation for Pretrained Transformers?
Shayne Longpre, Yu Wang, Christopher DuBois
Task-agnostic forms of data augmentation have proven widely effective in computer vision, even on pretrained models. In NLP similar results are reported most commonly for low data…
The Common Pile v0.1: An 8TB Dataset of Public Domain and Openly Licensed Text
Nikhil Kandpal, Brian Lester, Colin Raffel +24
Large language models (LLMs) are typically trained on enormous quantities of unlicensed text, a practice that has led to scrutiny due to possible intellectual property infringement…
The Responsible Foundation Model Development Cheatsheet: A Review of Tools & Resources
Shayne Longpre, Stella Biderman, Alon Albalak +20
Foundation model development attracts a rapidly expanding body of contributors, scientists, and applications. To help shape responsible development practices, we introduce the Foun…
Data Authenticity, Consent, & Provenance for AI are all broken: what will it take to fix them?
Shayne Longpre, Robert Mahari, Naana Obeng-Marnu +5
New capabilities in foundation models are owed in large part to massive, widely-sourced, and under-documented training data collections. Existing practices in data collection have…
Prometheus: Inducing Fine-grained Evaluation Capability in Language Models
Seungone Kim, Jamin Shin, Yejin Cho +8
Recently, using a powerful proprietary Large Language Model (LLM) (e.g., GPT-4) as an evaluator for long-form responses has become the de facto standard. However, for practitioners…
Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model
Ahmet Ãstün, Viraat Aryabumi, Zheng-Xin Yong +14
Recent breakthroughs in large language models (LLMs) have centered around a handful of data-rich languages. What does it take to broaden access to breakthroughs beyond first-class…
The BigScience ROOTS Corpus: A 1.6TB Composite Multilingual Dataset
Hugo Laurençon, Lucile Saulnier, Thomas Wang +51
As language models grow ever larger, the need for large-scale high-quality text datasets has never been more pressing, especially in multilingual settings. The BigScience workshop,…
Open-Domain Question Answering Goes Conversational via Question Rewriting
Raviteja Anantha, Svitlana Vakulenko, Zhucheng Tu +3
We introduce a new dataset for Question Rewriting in Conversational Context (QReCC), which contains 14K conversations with 80K question-answer pairs. The task in QReCC is to find a…
A Pretrainer's Guide to Training Data: Measuring the Effects of Data Age, Domain Coverage, Quality, & Toxicity
Shayne Longpre, Gregory Yauney, Emily Reif +8
Pretraining is the preliminary and fundamental step in developing capable language models (LM). Despite this, pretraining data design is critically under-documented and often guide…
Pivot Through English: Reliably Answering Multilingual Questions without Document Retrieval
Ivan Montero, Shayne Longpre, Ni Lao +2
Existing methods for open-retrieval question answering in lower resource languages (LRLs) lag significantly behind English. They not only suffer from the shortcomings of non-Englis…
The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI
Shayne Longpre, Robert Mahari, Anthony Chen +14
The race to train language models on vast, diverse, and inconsistently documented datasets has raised pressing concerns about the legal and ethical risks for practitioners. To reme…
Scaling Instruction-Finetuned Language Models
Hyung Won Chung, Le Hou, Shayne Longpre +32
Finetuning language models on a collection of datasets phrased as instructions has been shown to improve model performance and generalization to unseen tasks. In this paper we expl…
On the Transferability of Minimal Prediction Preserving Inputs in Question Answering
Shayne Longpre, Yi Lu, Christopher DuBois
Recent work (Feng et al., 2018) establishes the presence of short, uninterpretable input fragments that yield high confidence and accuracy in neural models. We refer to these as Mi…
Combining Compressions for Multiplicative Size Scaling on Natural Language Tasks
Rajiv Movva, Jinhao Lei, Shayne Longpre +2
Quantization, knowledge distillation, and magnitude pruning are among the most popular methods for neural network compression in NLP. Independently, these methods reduce model size…
Active Learning Over Multiple Domains in Natural Language Tasks
Shayne Longpre, Julia Reisler, Edward Greg Huang +4
Studies of active learning traditionally assume the target and source data stem from a single domain. However, in realistic applications, practitioners often require active learnin…
The Leaderboard Illusion
Shivalika Singh, Yiyang Nan, Alex Wang +10
Measuring progress is fundamental to the advancement of any scientific field. As benchmarks play an increasingly central role, they also grow more susceptible to distortion. Chatbo…
Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual Evaluation
Shivalika Singh, Angelika Romanou, Clémentine Fourrier +21
Cultural biases in multilingual datasets pose significant challenges for their effectiveness as global benchmarks. These biases stem not only from differences in language but also…
International Scientific Report on the Safety of Advanced AI (Interim Report)
Yoshua Bengio, Sören Mindermann, Daniel Privitera +41
This is the interim publication of the first International Scientific Report on the Safety of Advanced AI. The report synthesises the scientific understanding of general-purpose AI…
On the Societal Impact of Open Foundation Models
Sayash Kapoor, Rishi Bommasani, Kevin Klyman +22
Foundation models are powerful technologies: how they are released publicly directly shapes their societal impact. In this position paper, we focus on open foundation models, defin…
The 2024 Foundation Model Transparency Index
Rishi Bommasani, Kevin Klyman, Sayash Kapoor +4
Foundation models are increasingly consequential yet extremely opaque. To characterize the status quo, the Foundation Model Transparency Index (FMTI) was launched in October 2023 t…
FLARE-AI: Flaw Reporting for AI
Shayne Longpre, Elaine Zhu, Carson Ezell +15
Flaw reporting for deployed AI systems is fundamental to identifying system failures and improving AI safety. Yet the AI reporting ecosystem is fragmented: researchers who identify…
A Survey on Data Selection for Language Models
Alon Albalak, Yanai Elazar, Sang Michael Xie +11
A major factor in the recent success of large language models is the use of enormous and ever-growing text datasets for unsupervised pre-training. However, naively training a model…
An Exploration of Data Augmentation and Sampling Techniques for Domain-Agnostic Question Answering
Shayne Longpre, Yi Lu, Zhucheng Tu +1
To produce a domain-agnostic question answering model for the Machine Reading Question Answering (MRQA) 2019 Shared Task, we investigate the relative benefits of large pre-trained…
Future and AI-Ready Data Strategies: Response to DOC RFI on AI and Open Government Data Assets
Hamidah Oderinwale, Shayne Longpre
The following is a response to the US Department of Commerce's Request for Information (RFI) regarding AI and Open Government Data Assets. First, we commend the Department for its…
Foundation Model Transparency Reports
Rishi Bommasani, Kevin Klyman, Shayne Longpre +5
Foundation models are critical digital technologies with sweeping societal impact that necessitates transparency. To codify how foundation model developers should provide transpare…
The Foundation Model Transparency Index
Rishi Bommasani, Kevin Klyman, Shayne Longpre +5
Foundation models have rapidly permeated society, catalyzing a wave of generative AI applications spanning enterprise and consumer-facing contexts. While the societal impact of fou…
A Wrong Answer or a Wrong Question? An Intricate Relationship between Question Reformulation and Answer Selection in Conversational Question Answering
Svitlana Vakulenko, Shayne Longpre, Zhucheng Tu +1
The dependency between an adequate question formulation and correct answer selection is a very intriguing but still underexplored area. In this paper, we show that question rewriti…
Evaluating Entity Disambiguation and the Role of Popularity in Retrieval-Based NLP
Anthony Chen, Pallavi Gudipati, Shayne Longpre +2
Retrieval is a core component for open-domain NLP tasks. In open-domain tasks, multiple entities can share a name, making disambiguation an inherent yet under-explored problem. We…
Question Rewriting for Conversational Question Answering
Svitlana Vakulenko, Shayne Longpre, Zhucheng Tu +1
Conversational question answering (QA) requires the ability to correctly interpret a question in the context of previous conversation turns. We address the conversational QA task b…