LiteFrame: Efficient Vision Encoders Unlock Frame Scaling in Video LLMs Paper • 2605.17260 • Published 4 days ago • 22
Language-Switching Triggers Take a Latent Detour Through Language Models Paper • 2605.18646 • Published 3 days ago • 2
PEEK: Context Map as an Orientation Cache for Long-Context LLM Agents Paper • 2605.19932 • Published 1 day ago • 2
Where Does Authorship Signal Emerge in Encoder-Based Language Models? Paper • 2605.19908 • Published 1 day ago • 2
CopT: Contrastive On-Policy Thinking with Continuous Spaces for General and Agentic Reasoning Paper • 2605.20075 • Published 1 day ago • 2
Code-Guided Reasoning for Small Language Models: Evaluating Executable MCQA Scaffolds Paper • 2605.18827 • Published 9 days ago • 3
DocAtlas: Multilingual Document Understanding Across 80+ Languages Paper • 2605.12623 • Published 9 days ago • 2
Context Memorization for Efficient Long Context Generation Paper • 2605.18226 • Published 3 days ago • 3
WTF GENIUS PAPERS Collection Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 142 items • Updated about 4 hours ago • 27
Draft Less, Retrieve More: Hybrid Tree Construction for Speculative Decoding Paper • 2605.20104 • Published 1 day ago • 3
Overcoming Catastrophic Forgetting in Visual Continual Learning with Reinforcement Fine-Tuning Paper • 2605.09640 • Published 11 days ago • 5
Semantic Generative Tuning for Unified Multimodal Models Paper • 2605.18714 • Published 3 days ago • 7
WTF GENIUS PAPERS Collection Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 142 items • Updated about 4 hours ago • 27
CEPO: RLVR Self-Distillation using Contrastive Evidence Policy Optimization Paper • 2605.19436 • Published 1 day ago • 12
WTF GENIUS PAPERS Collection Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 142 items • Updated about 4 hours ago • 27
Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information Paper • 2605.11609 • Published 9 days ago • 60 • 2
Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information Paper • 2605.11609 • Published 9 days ago • 60