Junyan Lin
5 papers · 2024–2026 · 2 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓
🌍
Conference Polyglot · Bronze
🌉
Interdisciplinary Bridge · Bronze
🗺️
Taxonomy Completionist
(16)
Top co-authors
Research topics
Artificial Intelligence
(5)
Deep Learning
(5)
Core AI · Artificial Intelligence
(5)
Computer Vision
(3)
Machine Learning
(3)
Learning Types
(3)
Natural Language Processing
(2)
Core Methods
(2)
Models
(2)
Resources & Methods
(2)
Analysis
(1)
Application Areas · Machine Learning
(1)
Application Areas · Deep Learning
(1)
Architectures
(1)
Core AI · Computer Vision
(1)
Keywords
multimodal large language model
(3)
vision-language model
(2)
layer selection
(2)
attention mechanism
(1)
visual object tracking
(1)
chain-of-thought reasoning
(1)
multimodal learning
(1)
video understanding
(1)
visual representation
(1)
semantic alignment
(1)
feature fusion
(1)
token pruning
(1)
feature integration
(1)
vision-language integration
(1)
visual feature fusion
(1)
representation similarity
(1)
feature compression
(1)
connector selection
(1)
visual layer selection
(1)
cross-modal integration
(1)
Papers
Think-as-You-See: Streaming Chain-of-Thought Reasoning for Large Vision-Language Models
CVPR 2026
UTPTrack: Towards Simple and Unified Token Pruning for Visual Tracking
CVPR 2026
Multi-Layer Visual Feature Fusion in Multimodal LLMs: Methods, Analysis, and Best Practices
CVPR 2025
Multimodal Language Models See Better When They Look Shallower
EMNLP 2025
To Preserve or To Compress: An In-Depth Study of Connector Selection in Multimodal Large Language Models
EMNLP 2024