3 citations · 3 across the 17 of their papers we have counts for
18 papers
SFAD: Speculative Factuality-Aware Decoding
Guanqiao Chen, Di Wang, Lijie Hu
As one of the most critical challenges in large language models, contextual faithfulness directly determines their reliability in knowledge-intensive applications. This task is par…
Final Checkpoints Are Not Enough: Analyzing Latent Reasoning Faithfulness Along Training Trajectories
Hengyu Jin, Shu Yang, Di Wang
Latent reasoning performs multi-step inference in continuous hidden states, promising more compact and efficient reasoning. However, these opaque states raise a question of faithfu…
Mitigating Reasoning-Induced Misalignment via Safety-Direction Penalty
Yipeng Zhao, Qishun Yang, Shenzhe Zhu +2
Reasoning-Induced Misalignment, where fine-tuning on reasoning data containing no harmful content, including mathematics, code, and problem-solving with chain-of-thought traces can…
Visual Self-Fulfilling Alignment: Shaping Safety-Oriented Personas via Threat-Related Images
Qishun Yang, Shu Yang, Lijie Hu +1
Multimodal large language models (MLLMs) face safety misalignment, where visual inputs enable harmful outputs. To address this, existing methods require explicit safety labels or c…
Global Evolutionary Steering: Refining Activation Steering Control via Cross-Layer Consistency
Xinyan Jiang, Wenjing Yu, Di Wang +1
Activation engineering enables precise control over Large Language Models (LLMs) without the computational cost of fine-tuning. However, existing methods deriving vectors from stat…
Word Recovery in Large Language Models Enables Character-Level Tokenization Robustness
Zhipeng Yang, Shu Yang, Lijie Hu +1
Large language models (LLMs) trained with canonical tokenization exhibit surprising robustness to non-canonical inputs such as character-level tokenization, yet the mechanisms unde…
Beyond Scalars: Evaluating and Understanding LLM Reasoning via Geometric Progress and Stability
Xinyan Jiang, Ninghao Liu, Di Wang +1
Evaluating LLM reliability via scalar probabilities often fails to capture the structural dynamics of reasoning. We introduce TRACED, a framework that assesses reasoning quality th…
Towards Reasoning-Preserving Unlearning in Multimodal Large Language Models
Hongji Li, Junchi yao, Manjiang Yu +4
Machine unlearning aims to erase requested data from trained models without full retraining. For Reasoning Multimodal Large Language Models (RMLLMs), this is uniquely challenging:…
PIXEL: Adaptive Steering Via Position-wise Injection with eXact Estimated Levels under Subspace Calibration
Manjiang Yu, Hongji Li, Priyanka Singh +3
Reliable behavior control is central to deploying large language models (LLMs) on the web. Activation steering offers a tuning-free route to align attributes (e.g., truthfulness) t…
When Modalities Conflict: How Unimodal Reasoning Uncertainty Governs Preference Dynamics in MLLMs
Zhuoran Zhang, Tengyue Wang, Xilin Gong +4
Multimodal large language models (MLLMs) must resolve conflicts when different modalities provide contradictory information, a process we term modality following. Prior work measur…
Benchmarking and Mitigating Sycophancy in Medical Vision Language Models
Juangui Xu, Zikun Guo, Jingwei Lv +5
Visual language models (VLMs) have the potential to transform medical workflows. However, the deployment is limited by sycophancy. Despite this serious threat to patient safety, a…
When Truth Is Overridden: Uncovering the Internal Origins of Sycophancy in Large Language Models
Keyu Wang, Jin Li, Shu Yang +2
Large Language Models (LLMs) often exhibit sycophantic behavior, agreeing with user-stated opinions even when those contradict factual knowledge. While prior work has documented th…
The Compositional Architecture of Regret in Large Language Models
Xiangxiang Cui, Shu Yang, Tianjin Huang +3
Regret in Large Language Models refers to their explicit regret expression when presented with evidence contradicting their previously generated misinformation. Studying the regret…
Mitigating Behavioral Hallucination in Multimodal Large Language Models for Sequential Images
Liangliang You, Junchi Yao, Shu Yang +3
While multimodal large language models excel at various tasks, they still suffer from hallucinations, which limit their reliability and scalability for broader domain applications.…
Understanding and Mitigating Cross-lingual Privacy Leakage via Language-specific and Universal Privacy Neurons
Wenshuo Dong, Qingsong Yang, Shu Yang +5
Large Language Models (LLMs) trained on massive data capture rich information embedded in the training data. However, this also introduces the risk of privacy leakage, particularly…
Stable Vision Concept Transformers for Medical Diagnosis
Lijie Hu, Songning Lai, Yuan Hua +3
Transparency is a paramount concern in the medical field, prompting researchers to delve into the realm of explainable AI (XAI). Among these XAI methods, Concept Bottleneck Models…
Flattery in Motion: Benchmarking and Analyzing Sycophancy in Video-LLMs
Wenrui Zhou, Mohamed Hendy, Shu Yang +5
As video large language models (Video-LLMs) become increasingly integrated into real-world applications that demand grounded multimodal reasoning, ensuring their factual consistenc…
Understanding How Value Neurons Shape the Generation of Specified Values in LLMs
Yi Su, Jiayi Zhang, Shu Yang +3
Rapid integration of large language models (LLMs) into societal applications has intensified concerns about their alignment with universal ethical principles, as their internal val…