共计 140 篇文章
2026
KVSmooth： Mitigating Hallucination in Multi-modal Large Language Models through Key-Value Smoothing
HALLUCINATION-AWARE INTERMEDIATE REPRESENTATION EDIT IN LARGE VISION-LANGUAGE MODELS
J-space
Analyzing and Mitigating Object Hallucination： A Training Bias Perspective
More Tokens are Better but Not Necessary in Forensic Vision-Language Models
Mitigating the Object Hallucination of LVLM via Adaptive Factual-Guided Activation Editing
RFI： Rectified Flow Intervention for Mitigating Object Hallucination in Large Vision-Language Models
Toolformer：Language Models Can Teach Themselves to Use Tools
HalluZig：Hallucination Detection using Zigzag Persistence
Deeper is Not Always Better：Mitigating the Alignment Tax via Confident Layer Decoding