Thinking Inertia: LLMs Keep Thinking When Told Not To
Paper • 2610.11765 • Published • 217
Bingzheng Wei
Bingzheng
AI & ML interests
None yet
Recent Activity
upvoted a paper about 1 hour ago
Thinking Inertia: LLMs Keep Thinking When Told Not To upvoted a paper about 1 hour ago
GroundingPI: A Grounding Foundation Model towards Physical Intelligence with Visual Primitives upvoted a paper about 1 hour ago
S-Space: Exploring Spatial Workspace in Multimodal Language ModelsOrganizations
None yet
upvoted 3
papers about 1 hour ago
upvoted a paper about 20 hours ago
upvoted a paper 1 day ago
upvoted 2
papers 2 days ago
upvoted a paper 4 days ago
upvoted 12
papers 6 days ago
Rethinking Cross-Tokenizer On-Policy Distillation: From Alignment Coverage to Supervision Reliability
Paper • 2610.08448 • Published • 213
TRACE: Rollout-Guided Quantization-Aware Training for FP4 Reinforcement Learning of MoE Language Models
Paper • 2610.07767 • Published • 93
DuoMatching: Joint-Marginal Distribution Matching for Few-Step Video Generation
Paper • 2610.03543 • Published • 101
LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches
Paper • 2610.06647 • Published • 150
Does Learning Protein Folding Generalize to Broader Reasoning?
Paper • 2609.38879 • Published • 63
On-Policy or Off-Policy Learning? A Systematic Study of Distillation Dynamics
Paper • 2609.35259 • Published • 203
LoopVL: Recurrent Visual Intelligence
Paper • 2609.38426 • Published • 478
The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation
Paper • 2609.36484 • Published • 558
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents
Paper • 2609.39102 • Published • 655
Scaling Properties of Same-Family On-Policy Distillation
Paper • 2609.32722 • Published • 326
In-Context Learning for Robots: Methods and Applications
Paper • 2609.36012 • Published • 410
Omni-IO Skills: Harnessing Your Agent Omni-Native
Paper • 2609.31847 • Published • 451