221 citations · 496 across the 47 of their papers we have counts for
12 papers · 1 filter
The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning
Bill Yuchen Lin, Abhilasha Ravichander, Ximing Lu +5
The alignment tuning process of large language models (LLMs) typically involves instruction learning through supervised fine-tuning (SFT) and preference tuning via reinforcement le…
What Makes it Ok to Set a Fire? Iterative Self-distillation of Contexts and Rationales for Disambiguating Defeasible Social and Moral Situations
Kavel Rao, Liwei Jiang, Valentina Pyatkin +5
Moral or ethical judgments rely heavily on the specific contexts in which they occur. Understanding varying shades of defeasible contextualizations (i.e., additional information th…
The Generative AI Paradox: "What It Can Create, It May Not Understand"
Peter West, Ximing Lu, Nouha Dziri +11
The recent wave of generative AI has sparked unprecedented global attention, with both excitement and concern over potentially superhuman levels of artificial intelligence: models…
Phenomenal Yet Puzzling: Testing Inductive Reasoning Capabilities of Language Models with Hypothesis Refinement
Linlu Qiu, Liwei Jiang, Ximing Lu +8
The ability to derive underlying principles from a handful of observations and then generalize to novel situations -- known as inductive reasoning -- is central to human intelligen…
Value Kaleidoscope: Engaging AI with Pluralistic Human Values, Rights, and Duties
Taylor Sorensen, Liwei Jiang, Jena Hwang +10
Human values are crucial to human decision-making. Value pluralism is the view that multiple correct values may be held in tension with one another (e.g., when considering lying to…
Fine-Grained Human Feedback Gives Better Rewards for Language Model Training
Zeqiu Wu, Yushi Hu, Weijia Shi +6
Language models (LMs) often exhibit undesirable text generation behaviors, including generating false, toxic, or irrelevant outputs. Reinforcement learning from human feedback (RLH…
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…
Inference-Time Policy Adapters (IPA): Tailoring Extreme-Scale LMs without Fine-tuning
Ximing Lu, Faeze Brahman, Peter West +14
While extreme-scale language models have demonstrated exceptional performance on a variety of language tasks, the degree of control over these language models through pure promptin…
Evaluating Open-Domain Question Answering in the Era of Large Language Models
Ehsan Kamalloo, Nouha Dziri, Charles L. A. Clarke +1
Lexical matching remains the de facto evaluation method for open-domain question answering (QA). Unfortunately, lexical matching fails completely when a plausible candidate answer…
Elastic Weight Removal for Faithful and Abstractive Dialogue Generation
Nico Daheim, Nouha Dziri, Mrinmaya Sachan +2
Ideally, dialogue systems should generate responses that are faithful to the knowledge contained in relevant documents. However, many models generate hallucinated responses instead…
Self-Refine: Iterative Refinement with Self-Feedback
Aman Madaan, Niket Tandon, Prakhar Gupta +13
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…
CHAMPAGNE: Learning Real-world Conversation from Large-Scale Web Videos
Seungju Han, Jack Hessel, Nouha Dziri +2
Visual information is central to conversation: body gestures and physical behaviour, for example, contribute to meaning that transcends words alone. To date, however, most neural c…