6 papers
The Illusion of Stochasticity in LLMs
Xiangming Gu, Soham De, Michalis Titsias +3
In this work, we demonstrate that reliable stochastic sampling is a fundamental yet unfulfilled requirement for Large Language Models (LLMs) operating as agents. Agentic systems ar…
Unlocking Large Audio-Language Models for Interactive Language Learning
Hongfu Liu, Zhouying Cui, Xiangming Gu +1
Achieving pronunciation proficiency in a second language (L2) remains a challenge, despite the development of Computer-Assisted Pronunciation Training (CAPT) systems. Traditional C…
SkyLadder: Better and Faster Pretraining via Context Window Scheduling
Tongyao Zhu, Qian Liu, Haonan Wang +4
Recent advancements in LLM pretraining have featured ever-expanding context windows to process longer sequences. However, our pilot study reveals that models pretrained with shorte…
When Attention Sink Emerges in Language Models: An Empirical View
Xiangming Gu, Tianyu Pang, Chao Du +5
Language Models (LMs) assign significant attention to the first token, even if it is not semantically important, which is known as attention sink. This phenomenon has been widely a…
On Calibration of LLM-based Guard Models for Reliable Content Moderation
Hongfu Liu, Hengguan Huang, Xiangming Gu +2
Large language models (LLMs) pose significant risks due to the potential for generating harmful content or users attempting to evade guardrails. Existing studies have developed LLM…
On Memorization in Diffusion Models
Xiangming Gu, Chao Du, Tianyu Pang +3
Due to their capacity to generate novel and high-quality samples, diffusion models have attracted significant research interest in recent years. Notably, the typical training objec…