MemEIC: A Step Toward Continual and Compositional Knowledge Editing
"LVLMs" 相关文章
Process Reward Models for Sentence-Level Verification of LVLM Radiology Reports
CityRiSE: Reasoning Urban Socio-Economic Status in Vision-Language Models via Reinforcement Learning
Why LVLMs Are More Prone to Hallucinations in Longer Responses: The Role of Context
Learning to Detect Unknown Jailbreak Attacks in Large Vision-Language Models
PatentVision: A multimodal method for drafting patent applications
Effective and Stealthy One-Shot Jailbreaks on Deployed Mobile Vision-Language Agents
ChainMPQ: Interleaved Text-Image Reasoning Chains for Mitigating Relation Hallucinations
ChainMPQ: Interleaved Text-Image Reasoning Chains for Mitigating Relation Hallucinations
Multimodal Carotid Risk Stratification with Large Vision-Language Models: Benchmarking, Fine-Tuning, and Clinical Insights
INS-MMBench: A Comprehensive Benchmark for Evaluating LVLMs' Performance in Insurance
MAP: Mitigating Hallucinations in Large Vision-Language Models with Map-Level Attention Processing
Self-Aware Safety Augmentation: Leveraging Internal Semantic Understanding to Enhance Safety in Vision-Language Models
紫东太初开源视觉神经增强方法，即插即用终结多模态幻觉 | ACL 2025
入选ICML 2025！哈佛医学院等推出全球首个HIE领域临床思维图谱模型，神经认知结果预测任务上性能提升15%
入选ICML 2025！哈佛医学院等推出全球首个HIE领域临床思维图谱模型，神经认知结果预测任务上性能提升15%
让视觉语言模型像o3一样动手搜索、写代码！Visual ARFT实现多模态智能体能力
Teaching AI to Give Better Video Critiques
DeepSeek-R1的风吹到了多模态，Visual-RFT发布，视觉任务性能飙升20%
细粒度对齐无需仔细标注了！淘天提出视觉锚定奖励，自我校准实现多模态对齐
