565 citations · 1.3k across the 50 of their papers we have counts for
9 papers · 1 filter
NovaCOMET: Open Commonsense Foundation Models with Symbolic Knowledge Distillation
Peter West, Ronan Le Bras, Taylor Sorensen +8
We present NovaCOMET, an open commonsense knowledge model, that combines the best aspects of knowledge and general task models. Compared to previous knowledge models, NovaCOMET all…
MacGyver: Are Large Language Models Creative Problem Solvers?
Yufei Tian, Abhilasha Ravichander, Lianhui Qin +6
We explore the creative problem-solving capabilities of modern LLMs in a novel constrained setting. To this end, we create MACGYVER, an automatically generated dataset consisting o…
FANToM: A Benchmark for Stress-testing Machine Theory of Mind in Interactions
Hyunwoo Kim, Melanie Sclar, Xuhui Zhou +4
Theory of mind (ToM) evaluations currently focus on testing models using passive narratives that inherently lack interactivity. We introduce FANToM, a new benchmark designed to str…
Commonsense Knowledge Transfer for Pre-trained Language Models
Wangchunshu Zhou, Ronan Le Bras, Yejin Choi
Despite serving as the foundation models for a wide range of NLP benchmarks, pre-trained language models have shown limited capabilities of acquiring implicit commonsense knowledge…
Modular Transformers: Compressing Transformers into Modularized Layers for Flexible Efficient Inference
Wangchunshu Zhou, Ronan Le Bras, Yejin Choi
Pre-trained Transformer models like T5 and BART have advanced the state of the art on a wide range of text generation tasks. Compressing these models into smaller ones has become c…
NLPositionality: Characterizing Design Biases of Datasets and Models
Sebastin Santy, Jenny T. Liang, Ronan Le Bras +2
Design biases in NLP systems, such as performance differences for different populations, often stem from their creator's positionality, i.e., views and lived experiences shaped by…
From Dogwhistles to Bullhorns: Unveiling Coded Rhetoric with Language Models
Julia Mendelsohn, Ronan Le Bras, Yejin Choi +1
Dogwhistles are coded expressions that simultaneously convey one meaning to a broad audience and a second one, often hateful or provocative, to a narrow in-group; they are deployed…
Faith and Fate: Limits of Transformers on Compositionality
Nouha Dziri, Ximing Lu, Melanie Sclar +13
Transformer large language models (LLMs) have sparked admiration for their exceptional performance on tasks that demand intricate multi-step reasoning. Yet, these models simultaneo…
Leftover Lunch: Advantage-based Offline Reinforcement Learning for Language Models
Ashutosh Baheti, Ximing Lu, Faeze Brahman +3
Reinforcement Learning with Human Feedback (RLHF) is the most prominent method for Language Model (LM) alignment. However, RLHF is an unstable and data-hungry process that continua…