7 citations · 11 across the 12 of their papers we have counts for
15 papers
Sparse Autoencoders Are a Shared Coordinate System for Symbolic Matching
Xiaocong Yang, Xuanyu Meng, Yuji Zhang +2
Large language models (LLMs) hold broad knowledge in one set of weights, but lack interpretability and trustworthy reasoning, especially in high-stakes use cases. A neuro-symbolic…
Popular Knowledge Propagates More Errors in LLM Knowledge Updating
Yuji Zhang, Weibing Wang, Cheng Qian +5
Updating a language model's knowledge through fine-tuning is essential for keeping its outputs current, yet can also induce factual forgetting and new hallucinations. Prior work sh…
Benchmarking Multi-turn Medical Diagnosis: Hold, Lure, and Self-Correction
Jinrui Fang, Runhan Chen, Xu Yang +9
Large language models (LLMs) achieve high accuracy in medical diagnosis when all clinical information is provided in a single turn, yet how they behave under multi-turn evidence ac…
Wired for Overconfidence: A Mechanistic Perspective on Inflated Verbalized Confidence in LLMs
Tianyi Zhao, Yinhan He, Wendy Zheng +2
Large language models are often not just wrong, but \emph{confidently wrong}: when they produce factually incorrect answers, they tend to verbalize overly high confidence rather th…
Current Agents Fail to Leverage World Model as Tool for Foresight
Cheng Qian, Emre Can Acikgoz, Bingxuan Li +8
Agents built on vision-language models increasingly face tasks that demand anticipating future states rather than relying on short-horizon reasoning. Generative world models offer…
EMCompress: Video-LLMs with Endomorphic Multimodal Compression
Zheyu Fan, Jiateng Liu, Yuji Zhang +4
Video-LLMs face a fundamental tension in long-video reasoning: static, sparse frame sampling either dilutes evidence across task-irrelevant segments at significant cost or misses f…
Atomic Reasoning for Scientific Table Claim Verification
Yuji Zhang, Qingyun Wang, Cheng Qian +7
Scientific texts often convey authority due to their technical language and complex data. However, this complexity can sometimes lead to the spread of misinformation. Non-experts a…
ModelingAgent: Bridging LLMs and Mathematical Modeling for Real-World Challenges
Cheng Qian, Hongyi Du, Hongru Wang +6
Recent progress in large language models (LLMs) has enabled substantial advances in solving mathematical problems. However, existing benchmarks often fail to reflect the complexity…
TAMA: A Human-AI Collaborative Thematic Analysis Framework Using Multi-Agent LLMs for Clinical Interviews
Huimin Xu, Seungjun Yi, Terence Lim +9
Thematic analysis (TA) is a widely used qualitative approach for uncovering latent meanings in unstructured text data. TA provides valuable insights in healthcare but is resource-i…
The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination
Yuji Zhang, Sha Li, Cheng Qian +8
Hallucination is a persistent challenge in large language models (LLMs), where even with rigorous quality control, models often generate distorted facts. This paradox, in which err…
SafeSwitch: Steering Unsafe LLM Behavior via Internal Activation Signals
Peixuan Han, Cheng Qian, Xiusi Chen +3
Large language models (LLMs) exhibit exceptional capabilities across various tasks but also pose risks by generating harmful content. Existing safety mechanisms, while improving mo…
EscapeBench: Towards Advancing Creative Intelligence of Language Model Agents
Cheng Qian, Peixuan Han, Qinyu Luo +9
Language model agents excel in long-session planning and reasoning, but existing benchmarks primarily focus on goal-oriented tasks with explicit objectives, neglecting creative ada…
A Survey on the Honesty of Large Language Models
Siheng Li, Cheng Yang, Taiqiang Wu +12
Honesty is a fundamental principle for aligning large language models (LLMs) with human values, requiring these models to recognize what they know and don't know and be able to fai…
Knowledge Overshadowing Causes Amalgamated Hallucination in Large Language Models
Yuji Zhang, Sha Li, Jiateng Liu +5
Hallucination is often regarded as a major impediment for using large language models (LLMs), especially for knowledge-intensive tasks. Even when the training corpus consists solel…
EVEDIT: Event-based Knowledge Editing with Deductive Editing Boundaries
Jiateng Liu, Pengfei Yu, Yuji Zhang +3
The dynamic nature of real-world information necessitates efficient knowledge editing (KE) in large language models (LLMs) for knowledge updating. However, current KE approaches, w…