7 papers
Multi-Turn Reflective Masking Elicits Reasoning in Mask Diffusion Models
Yanming Zhang, Yihan Bian, Jingyuan Qi +3
While reasoning on autoregressive (AR) models is often performed by chain-of-thought reasoning and reflection, their refinement of previous outputs still relies on fully sequential…
Sandboxed Coding Agents are Competitive Omni-modal Task Solvers
Dongping Chen, Xuanao Huang, Zhihan Hu +3
As multimodal LLMs increasingly target video and audio, it is often assumed that such tasks require native omnimodal models. We show that this is not always the case: coding agents…
How Language Models Process Negation
Zhejian Zhou, Tianyi Zhou, Robin Jia +1
We study how Large Language Models (LLMs) process negation mechanistically. First, we establish that even though open-weight models often provide wrong answers to questions involvi…
AI, Take the Wheel: What Drives Delegation and Trust in Human-Computer Cooperative Question Answering?
Maharshi Gor, Yoo Yeon Sung, Yu Hou +4
AI systems are fallible, and humans can make mistakes in deciding whether to trust AI over their own judgment. Thus, improving human-AI collaboration requires understanding when, w…
Quantifying the Gap between Understanding and Generation within Unified Multimodal Models
Chenlong Wang, Yuhang Chen, Zhihan Hu +4
Recent advances in unified multimodal models (UMM) have demonstrated remarkable progress in both understanding and generation tasks. However, whether these two capabilities are gen…
Optimizing Length Compression in Large Reasoning Models
Zhengxiang Cheng, Dongping Chen, Mingyang Fu +1
Large Reasoning Models (LRMs) have achieved remarkable success, yet they often suffer from producing unnecessary and verbose reasoning chains. We identify a core aspect of this iss…