12 papers
Human Cognitive Benchmarks Reveal Foundational Visual Gaps in MLLMs
Jen-Tse Huang, Dasen Dai, Jen-Yuan Huang +7
Humans develop perception through a bottom-up hierarchy: from basic primitives and Gestalt principles to high-level semantics. In contrast, current Multimodal Large Language Models…
Identifying the Achilles' Heel: An Iterative Method for Dynamically Uncovering Factual Errors in Large Language Models
Wenxuan Wang, Yuk-Kit Chan, Zixuan Ling +7
Large Language Models (LLMs) like ChatGPT are foundational in various applications due to their extensive knowledge from pre-training and fine-tuning. Despite this, they are prone…
Curing Miracle Steps in LLM Mathematical Reasoning with Rubric Rewards
Youliang Yuan, Qiuyang Mang, Jingbang Chen +7
In this paper, we observe that current models are susceptible to reward hacking, leading to a substantial overestimation of a model's reasoning ability. This is evidenced by a high…
DeepCompress: A Dual Reward Strategy for Dynamically Exploring and Compressing Reasoning Chains
Tian Liang, Wenxiang Jiao, Zhiwei He +3
Large Reasoning Models (LRMs) have demonstrated impressive capabilities but suffer from cognitive inefficiencies like "overthinking" simple problems and "underthinking" complex one…
Towards Evaluating Proactive Risk Awareness of Multimodal Language Models
Youliang Yuan, Wenxiang Jiao, Yuejin Xie +5
Human safety awareness gaps often prevent the timely recognition of everyday risks. In solving this problem, a proactive safety artificial intelligence (AI) system would work bette…
Benchmarking Chinese Commonsense Reasoning with a Multi-hop Reasoning Perspective
Wangjie You, Xusheng Wang, Xing Wang +4
While Large Language Models (LLMs) have demonstrated advanced reasoning capabilities, their comprehensive evaluation in general Chinese-language contexts remains understudied. To b…