Is CLIP Cross-Eyed? Revealing and Mitigating Center Bias in the CLIP Family
Paper • 2604.05971 • Published
AI & ML interests
We do research on Trustworthy, Multilingual, and Multimodal NLP/LLMs.
Recent Activity
View all activity
Papers
Not Every Mention Is a Claim: Rethinking Object Hallucination Detection in Vision-Language Models
The First Drop of Ink: Nonlinear Impact of Distracting Information in Long-Context Reasoning
kuanhaoh
authored 8
papers 15 days ago
Steering Vector Fields for Context-Aware Inference-Time Control in Large Language Models
Paper • 2602.01654 • Published
TRACES: Proactive Safety Auditing for Multi-Turn LLM Agents via Trajectory-State Modeling
Paper • 2605.27690 • Published
Language Steering for Multilingual In-Context Learning
Paper • 2602.02326 • Published
TwinICL: Diagnosing Multimodal In-Context Learning through Paired Counterfactuals
Paper • 2609.15028 • Published
ChronoCulture: Evaluating Cultural Awareness Across Time in Large Language Models
Paper • 2610.20472 • Published
Not Every Mention Is a Claim: Rethinking Object Hallucination Detection in Vision-Language Models
Paper • 2610.23610 • Published • 2
Are We Really Measuring Forgetting? Pitfalls in Evaluating Machine Unlearning for Multimodal LLMs
Paper • 2610.19613 • Published
JustinHsu99
authored 5
papers 18 days ago
MedPlan: A Two-Stage RAG-Based System for Personalized Medical Plan Generation
Paper • 2503.17900 • Published • 1
MedAction: Towards Active Multi-turn Clinical Diagnostic LLMs
Paper • 2605.07305 • Published • 1
Not Every Mention Is a Claim: Rethinking Object Hallucination Detection in Vision-Language Models
Paper • 2610.23610 • Published • 2
Do LLMs Understand Sequential Structure? A Controlled Study of Inference and Generation
Paper • 2610.04977 • Published • 13
DAT: Dynamic Alpha Tuning for Hybrid Retrieval in Retrieval-Augmented Generation
Paper • 2503.23013 • Published • 1
kuanhaoh
authored 4
papers 21 days ago
The First Drop of Ink: Nonlinear Impact of Distracting Information in Long-Context Reasoning
Paper • 2605.10828 • Published • 2
When Correct Demonstrations Hurt: Rethinking the Role of Exemplars in In-Context Learning
Paper • 2605.26350 • Published
Pop-Up Distractions Reveal Bag-of-Events Behavior in Video Large Language Models
Paper • 2605.27101 • Published
ConstructCIE: A Dataset for Extracting Causal Information from Construction Accident Narratives
Paper • 2608.06495 • Published
MuhanGao
updated a
dataset about 1 month ago
zihan-xue
updated a
dataset about 2 months ago
SussyCat
updated a
dataset 2 months ago