Sharon Li

How Do Transformers Learn to Associate Tokens: Gradient Leading Terms Bring Mechanistic Interpretability

KnowledgeSmith: Uncovering Knowledge Updating in LLMs with Model Editing and Unlearning

LUMINA: Detecting Hallucinations in RAG System with Context–Knowledge Signals

Hybrid Reinforcement: when reward is sparse, better to be dense

Beyond In-Domain Detection: SpikeScore for Cross-Domain Hallucination Detection

LH-DECEPTION: Simulating and Understanding LLM Deceptive Behaviors in Long-Horizon Interactions

Understanding Language Prior of LVLMs by Contrasting Chain-of-Embedding

General Exploratory Bonus for Optimistic Exploration in RLHF

Towards Interpretability Without Sacrifice: Faithful Dense Layer Decomposition with Mixture of Decoders

MetaMind: Modeling Human Social Thoughts with Metacognitive Multi-Agent Systems

GeoRanker: Distance-Aware Ranking for Worldwide Image Geolocalization

Limited Preference Data? Learning Better Reward Model with Latent Space Synthesis

Harnessing Feature Resonance under Arbitrary Target Alignment for Out-of-Distribution Node Detection

Debate or Vote: Which Yields Better Decisions in Multi-Agent Large Language Models?

Your Pre-trained LLM is Secretly an Unsupervised Confidence Calibrator

Visual Instruction Bottleneck Tuning

GLSim: Detecting Object Hallucinations in LVLMs via Global-Local Similarity

Can DPO Learn Diverse Human Values? A Theoretical Scaling Law