Large Language Model
End-To-End Causal Effect Estimation from Unstructured Natural Language Data
Work done during internship at Meta AI. 38th Conference on Neural Information Processing Systems (NeurIPS 2024).
Learning to Discuss Strategically: A Case Study on One Night Ultimate Werewolf
As a variant of the famous communication game Werewolf, One Night Ultimate W erewolf (ONUW) requires players to develop strategic discussion policies due to the potential role changes that increase the uncertainty and complexity of the game.
Many-Shot In-Context Learning Rishabh Agarwal, A vi Singh
LLMs excel at few-shot in-context learning (ICL) - learning from a few input-output examples ("shots") provided in context at inference, without any weight updates.
Few-shot Algorithms for Consistent Neural Decoding (FALCON) Benchmark Brianna M. Karpowicz 1,2 Joel Y e
However, progress is limited by a lack of standardized datasets and comparison metrics, causing methods to be compared in an ad hoc manner.
CoMat: Aligning Text-to-Image Diffusion Model with Image-to-Text Concept Matching
Diffusion models have demonstrated great success in the field of text-to-image generation. However, alleviating the misalignment between the text prompts and images is still challenging. We break down the problem into two causes: concept ignorance and concept mismapping.
