Why Is Spatial Reasoning Hard for VLMs? An Attention Mechanism Perspective on Focus Areas
Paper • 2503.01773 • Published
AI & ML interests
None defined yet.
Recent Activity
View all activity
ManlingLi
authored 20
papers 6 days ago
Bring Reason to Vision: Understanding Perception and Reasoning through Model Merging
Paper • 2505.05464 • Published • 11
RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learning
Paper • 2504.20073 • Published • 12
Non-Sequential Graph Script Induction via Multimedia Grounding
Paper • 2305.17542 • Published • 1
The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination
Paper • 2502.16143 • Published • 6
Re-thinking Temporal Search for Long-Form Video Understanding
Paper • 2504.02259 • Published • 1
HourVideo: 1-Hour Video-Language Understanding
Paper • 2411.04998 • Published • 1
IKEA Manuals at Work: 4D Grounding of Assembly Instructions on Internet Videos
Paper • 2411.11409 • Published
COVID-19 Literature Knowledge Graph Construction and Drug Repurposing Report Generation
Paper • 2007.00576 • Published • 1
Text-Based Reasoning About Vector Graphics
Paper • 2404.06479 • Published
Multimedia Generative Script Learning for Task Planning
Paper • 2208.12306 • Published • 2
Conv-CoA: Improving Open-domain Question Answering in Large Language Models via Conversational Chain-of-Action
Paper • 2405.17822 • Published
Knowledge Overshadowing Causes Amalgamated Hallucination in Large Language Models
Paper • 2407.08039 • Published • 1
Word Embeddings Are Steers for Language Models
Paper • 2305.12798 • Published
MentalArena: Self-play Training of Language Models for Diagnosis and Treatment of Mental Health Disorders
Paper • 2410.06845 • Published • 5
Embodied Agent Interface: Benchmarking LLMs for Embodied Decision Making
Paper • 2410.07166 • Published • 3
Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models
Paper • 2403.17359 • Published
LayoutVLM: Differentiable Optimization of 3D Layout via Vision-Language Models
Paper • 2412.02193 • Published • 1
Spatial Mental Modeling from Limited Views
Paper • 2506.21458 • Published • 13
A Simple "Try Again" Can Elicit Multi-Turn LLM Reasoning
Paper • 2507.14295 • Published • 14