5 papers · 1 filter
Faster but Different: Diagnosing and Controlling Content Drift in Accelerated Multimodal Diffusion Language Models
Yaoxuan Dou, Yang Shu
Training-free acceleration makes diffusion-based multimodal large language models (dMLLMs) more deployable, but it may silently change generated content. We study this serving-time…
Gavel: Agent Meets Checklist for Evaluating LLMs on Long-Context Legal Summarization
Yao Dou, Benjamin Mamut, Wei Xu
Large language models (LLMs) now support contexts of up to 1M tokens, but their strengths and weaknesses on complex long-context tasks remain unclear. To study this, we focus on mu…
Localizing Prompt Ambiguity in Large Language Models with Probe-Targeted Attribution
Govind Ramesh, Yao Dou, Wei Xu
Prompt ambiguity is a common source of failure in large language models, but is difficult to localize because it is a latent property of the prompt, while existing attribution meth…
Evaluating LLMs on Chinese Idiom Translation
Cai Yang, Yao Dou, David Heineman +2
Idioms, whose figurative meanings usually differ from their literal interpretations, are common in everyday language, especially in Chinese, where they often contain historical ref…
Improving Minimum Bayes Risk Decoding with Multi-Prompt
David Heineman, Yao Dou, Wei Xu
While instruction fine-tuned LLMs are effective text generators, sensitivity to prompt construction makes performance unstable and sub-optimal in practice. Relying on a single "bes…