Research - Papers
Explore a selection of our published work on a variety of key research challenges in AI.
Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research
Information about pretraining corpora used to train the current best-performing language models is seldom discussed: commercial models rarely detail their data, and even open models are often…
Catwalk: A Unified Language Model Evaluation Framework for Many Datasets
The success of large language models has shifted the evaluation paradigms in natural language processing (NLP). The community's interest has drifted towards comparing NLP models across many tasks,…
IfQA: A Dataset for Open-domain Question Answering under Counterfactual Presuppositions
Although counterfactual reasoning is a fundamental aspect of intelligence, the lack of large-scale counterfactual open-domain question-answering (QA) benchmarks makes it difficult to evaluate and…
Self-Refine: Iterative Refinement with Self-Feedback
Like humans, large language models (LLMs) do not always generate the best output on their first try. Motivated by how humans refine their written text, we introduce Self-Refine, an approach for…
A Logic for Expressing Log-Precision Transformers
One way to interpret the reasoning power of transformer-based language models is to describe the types of logical rules they can resolve over some input text. Recently, Chiang et al. (2023) showed…
How Far Can Camels Go? Exploring the State of Instruction Tuning on Open Resources
In this work we explore recent advances in instruction-tuning language models on a range of open instruction-following datasets. Despite recent claims that open models can be on par with…
Editing Common Sense in Transformers
Editing model parameters directly in Transformers makes updating open-source transformer-based models possible without re-training. However, these editing methods have only been evaluated on…
Increasing Probability Mass on Answer Choices Does Not Always Improve Accuracy
When pretrained language models (LMs) are applied to discriminative tasks such as multiple-choice questions, they place probability mass on vocabulary tokens that aren't among the given answer…
Language Models with Rationality
While large language models (LLMs) are proficient at question-answering (QA), the dependencies between their answers and other "beliefs" they may have about the world are typically unstated, and may…
What Makes it Ok to Set a Fire? Iterative Self-distillation of Contexts and Rationales for Disambiguating Defeasible Social and Moral Situations
Moral or ethical judgments rely heavily on the specific contexts in which they occur. Understanding varying shades of defeasible contextualizations (i.e., additional information that strengthens or…