Papers

• REFACT: Adaptive Fact Restatement for Compact and Faithful Chain-of-Thought Reasoning (2026)
  Zhensheng Jin et al.
  5.01
• ACID: Action Consistency via Inverse Dynamics for Planning with World Models (2026)
  Gawon Seo et al.
  2.00
• RePro: Training Language Models to Faithfully Recycle the Web for Pretraining (2025)
  Zichun Yu et al.
  1.50
• Learning to Reason as Action Abstractions with Scalable Mid-Training RL (2025)
  Shenao Zhang et al.
  1.44
• A Token is Worth over 1,000 Tokens: Efficient Knowledge Distillation through Low-Rank Clone (2025)
  Jitai Hao et al.
  1.28
• Plan and Budget: Effective and Efficient Test-Time Scaling on Large Language Model Reasoning (2025)
  Junhong Lin et al.
  1.28
• DeepPrune: Parallel Scaling without Inter-trace Redundancy (2025)
  Shangqing Tu et al.
  1.28
• ReCode: Unify Plan and Action for Universal Granularity Control (2025)
  Zhaoyang Yu et al.
  1.28
• Rethinking Optimal Verification Granularity for Compute-Efficient Test-Time Scaling (2025)
  Hao Mark Chen et al.
  1.22
• R2R: Efficiently Navigating Divergent Reasoning Paths with Small-Large Model Token Routing (2025)
  Tianyu Fu et al.
  1.22
• PseudoReasoner: Leveraging Pseudo Labels for Commonsense Knowledge Base Population (2022)
  Tianqing Fang et al.
  —
• Call Me When Necessary: LLMs can Efficiently and Faithfully Reason over Structured Environments (2024)
  Sitao Cheng et al.
  —