An open API service indexing awesome lists of open source software.

Awesome-Efficient-Reasoning

Paper list for Efficient Reasoning.
https://github.com/hemingkx/Awesome-Efficient-Reasoning

Last synced: 5 days ago
JSON representation

  • Papers

    • Long-to-Short Chain-of-Thought

      • [pdf - orange) ![](https://img.shields.io/badge/Early_Exit-green)
      • [pdf - orange) ![](https://img.shields.io/badge/BudgetThinker-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/DRQA-blue)
      • [pdf - -findings-orange) ![](https://img.shields.io/badge/CAC--CoT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/ThinkDial-blue) ![](https://img.shields.io/badge/gpt--oss--style-green)
      • [pdf - orange) ![](https://img.shields.io/badge/AGPO-blue) ![](https://img.shields.io/badge/length--based_reward-green)
      • [pdf - to-Short-via-Model-Merging)], 2025.03. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Model_Merging-green)
      • [pdf - orange) ![](https://img.shields.io/badge/ShorterBetter-blue) ![](https://img.shields.io/badge/Group--Relative_Length_Reward-green)
      • [pdf - orange) ![](https://img.shields.io/badge/AdaR1-blue)
      • [pdf - AI4Edu/LS-Mixture)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/LS--Mixture_SFT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/PI-blue)
      • [pdf - orange)
      • [pdf - Group/SEAL)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/SEAL-blue) ![](https://img.shields.io/badge/Steering-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Significance--aware_Length_Penalty-green)
      • [pdf - -findings-orange) ![](https://img.shields.io/badge/Prompting-green)
      • [pdf - orange) ![](https://img.shields.io/badge/In--context_Learning&SFT-green)
      • [pdf - -findings-orange) ![](https://img.shields.io/badge/NoWait-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Prompting-green)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - Embodied-AGI/BudgetGuidance)], 2025.06. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - R1)], 2025.06. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Steering-green)
      • [pdf - orange)
      • [pdf - NLP-Chang/ThinkPrune)], 2025.04. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/ThinkPrune-blue)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Elastic_Reasoning-blue) ![](https://img.shields.io/badge/Length_Control-green)
      • [pdf - orange) ![](https://img.shields.io/badge/S--GRPO-blue) ![](https://img.shields.io/badge/Early_Exit-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Length_Penalty-green)
      • [pdf - orange) ![](https://img.shields.io/badge/NoThinking-blue) ![](https://img.shields.io/badge/Prompt-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Length_Penalty-green)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Multilingual-green)
      • [pdf - nlp-lab/LengthReward)], 2025.06. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - Policy-Preference-Optimization)], 2025.07. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Steering-green)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Steering-green)
      • [pdf - -findings-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Multilingual-green)
      • [pdf - orange) ![](https://img.shields.io/badge/SelfBudgeter-blue) ![](https://img.shields.io/badge/Adaptive_Token_Budget-green)
      • [pdf - otimes-Short/)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Prune--on--Logic-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/FlashThink-blue) ![](https://img.shields.io/badge/Early_Exit-green)
      • [pdf - orange) ![](https://img.shields.io/badge/VeriThinker-blue) ![](https://img.shields.io/badge/CoTs_by_InstructLM-green)
      • [pdf - sg/AnytimeReasoner)], 2025.05. ![](https://img.shields.io/badge/NeurIPS2025-orange) ![](https://img.shields.io/badge/BRPO-blue) ![](https://img.shields.io/badge/Early_Exit-green)
      • [pdf - orange) ![](https://img.shields.io/badge/ThinkLess-blue) ![](https://img.shields.io/badge/Early_Exit-green)
      • [pdf - nlp/Laser)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Laser-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/ACPO-blue)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/ConciseRL-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/TrimR-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/ALP-blue) ![](https://img.shields.io/badge/Length_Control-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Length_Penalty-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Outline_by_InstructLM-green)
      • [pdf - orange) ![](https://img.shields.io/badge/TL;DR-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Early_Exit-green)
      • [pdf - Reasoning-Efficiency)], 2025.06. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/REO--RL-blue)
      • [pdf - paper/)], [[code](https://github.com/royeisen/reasoning_loading_bar)], 2025.06. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - Thought)], 2025.05. ![](https://img.shields.io/badge/NeurIPS2025-orange) ![](https://img.shields.io/badge/A*--Thought-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/DEER-blue) ![](https://img.shields.io/badge/Early_Exit-green)
      • [pdf - of-symbol-planning)], 2023.05. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Dynasor-blue)
      • [pdf - Pruner)], 2025.01. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/O1--Pruner-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Kimi_k1.5-blue)
      • [pdf - Labs/efficient-reasoning)], [[homepage](https://zanette-labs.github.io/efficient-reasoning/)], 2025.02. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Meta--Reasoner-blue)
      • [pdf - -findings-orange) ![](https://img.shields.io/badge/TALE-blue) ![](https://img.shields.io/badge/Prompt-green)
      • [pdf - concise-cot)], 2024.01. ![](https://img.shields.io/badge/FLLM2024-orange) ![](https://img.shields.io/badge/CCoT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/C3oT-blue)
      • [pdf - of-draft)], 2025.02. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/CoD-blue) ![](https://img.shields.io/badge/Prompt-green)
      • [pdf - CoT/compressed-cot)], 2025.03. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Prompt-green)
      • [pdf - orange) ![](https://img.shields.io/badge/SoT-blue) ![](https://img.shields.io/badge/Prompt-green)
      • [pdf - Valve)], 2025.02. ![](https://img.shields.io/badge/ACL2025-orange) ![](https://img.shields.io/badge/CoT--Valve-blue)
      • [pdf - reasoning)], 2025.02. ![](https://img.shields.io/badge/ACL2025--findings-orange)
      • [pdf - l3/l1)], [[homepage](https://cmu-l3.github.io/l1/)], 2025.03. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/L1-blue) ![](https://img.shields.io/badge/Length_Control-green)
      • [pdf - orange) ![](https://img.shields.io/badge/DAST-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/CGRS-blue)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/GFPO-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/SABER-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/VSRM-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/DR.SAF-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Length_Penalty-green)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/EDIT-blue)
      • [pdf - real/hbpo)], 2025.09. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/HBPO-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/ES--COT-blue)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/MI-blue)
      • [pdf - dev/ReasoningPathCompression)], 2025.05. ![](https://img.shields.io/badge/NeurIPS2025-orange) ![](https://img.shields.io/badge/KV_Cache_Pruning-green)
      • [pdf - orange) ![](https://img.shields.io/badge/MACC-blue)
      • [pdf - Enough-Think)], 2025.09. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/JET-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Early_Exit-green)
      • [pdf - KEG/siri)], 2025.10. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/SIRI-blue)
    • Analysis

      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Over--thinking-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Over--thinking-green)
      • [pdf - lab/MiP-Overthinking)], 2025.04. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Over--thinking-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Long2Short_Inconsistency-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Reasoning_Steps-green)
      • [pdf - ov-file)], 2025.06. ![](https://img.shields.io/badge/ACL2025--Oral-orange)
      • [pdf - latent-cot)], 2025.07. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Over--thinking-green)
      • [pdf - USTC/LRM-plans-CoT)], 2025.06. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Reasoning_Length-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Length_Budget-green)
      • [pdf - Impact-of-Reasoning-Step-Length-on-Large-Language-Models)], 2024.01. ![](https://img.shields.io/badge/AC2024--findings-orange) ![](https://img.shields.io/badge/Reasoning_Step_Length-green)
      • [pdf - boundary)], 2024.10. ![](https://img.shields.io/badge/NIPS2024-orange) ![](https://img.shields.io/badge/Reasoning_Boundary-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Over--thinking-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Optimal_CoT_Length-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Over--thinking-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Under--thinking-green)
      • [pdf - orange)
      • [pdf - 32B)], 2025.03. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Over--thinking-green)
      • [pdf - try-matters)], 2025.10. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange)
    • Small Reasoning Models & CoT Distillation

      • [pdf - NLP/Distilling-CoT-Reasoning)], 2025.02. ![](https://img.shields.io/badge/ACL2025--findings-orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - -findings-orange)
      • [pdf - z/SCORE)], 2024.04. ![](https://img.shields.io/badge/ACL2024--findings-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/FDD-blue)
      • [pdf - orange)
      • [pdf - optimal-tts)], [[homepage](https://ryanliu112.github.io/compute-optimal-tts/)], 2025.02. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/TTS-blue)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Survey-green)
      • [pdf - -short-orange)
      • [pdf - wang/Tina)], 2025.04. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Tina-blue)
      • [pdf - orange)
      • [pdf - luo/SODE)], 2025.03. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/DLCoT-blue)
      • [pdf - nlp/simpleRL-reason)], 2025.03. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/SimpleRL--Zoo-blue) ![](https://img.shields.io/badge/Zero_RL_Training-green)
      • [pdf - rs)], 2025.03. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/TwT-blue)
      • [pdf - Model-Gap/Small-Model-Learnability-Gap)], [[homepage](https://small-model-gap.github.io/)], 2025.02. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - USTC/TFPI-temp)], 2025.09. ![](https://img.shields.io/badge/Arxiv-orange)
    • Speculative Decoding for CoT Efficiency

      • [pdf - orange) ![](https://img.shields.io/badge/SCoT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/SpecReason-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Speculative_Thinking-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/STAND-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/RSD-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/R2R-blue)
      • [pdf - ai-lab/LookaheadReasoning)], 2025.06. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/FastGRPO-blue)
      • [pdf - RL)], 2025.09. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/SPEC--RL-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/SpecExit-blue)
    • Survey

      • [pdf - Efficient-R1-style-LRMs)], 2025.07. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - Efficient-Reasoning-Models)], 2025.04. ![](https://img.shields.io/badge/TMLR2025-orange)
      • [pdf - llm-implicit-reasoning)], 2025.09. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange)
      • [pdf - art-projection/LatentCoT-Horizon/)], 2025.07. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange)
      • [pdf - Efficient-Reasoning-LLMs)], 2025.03. ![](https://img.shields.io/badge/TMLR2025-orange)
      • [pdf - orange)
      • [pdf - Efficient-Inference-for-LRMs)], 2025.03. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - Reasoning-Economy-Papers)], 2025.03. ![](https://img.shields.io/badge/Arxiv-orange)
    • Other Work

      • [pdf - orange) ![](https://img.shields.io/badge/M1-blue) ![](https://img.shields.io/badge/Mamba-green)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/System2-green)
      • [pdf - orange) ![](https://img.shields.io/badge/System2-green)
      • [pdf - orange) ![](https://img.shields.io/badge/R2--Reasoner-blue) ![](https://img.shields.io/badge/Model_Router-green)
      • [pdf - orange) ![](https://img.shields.io/badge/PENCIL-blue) ![](https://img.shields.io/badge/Intermediate_CoT_Reduction-green)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)
    • Efficient Training

      • [pdf - orange) ![](https://img.shields.io/badge/SPEED--RL-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/T--PPO-blue)
      • [pdf - lime/verl)], 2025.04. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/ADARFT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/VAPO-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/QFFT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/TreePO-blue)
      • [pdf - R1)], 2025.03. ![](https://img.shields.io/badge/ACL2025--industry-orange) ![](https://img.shields.io/badge/Light--R1-blue)
      • [pdf - ai-lab.github.io/GRESO/)], [[code](https://github.com/Infini-AI-Lab/GRESO)], 2025.06. ![](https://img.shields.io/badge/NeurIPS2025-orange) ![](https://img.shields.io/badge/Selective_Rollouts-green)
      • [pdf - cpu/Question-Free-Fine-Tuning)], 2025.06. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/QFFT-blue)
      • [pdf - PO)], 2025.05. ![](https://img.shields.io/badge/NeurIPS2025-orange) ![](https://img.shields.io/badge/A*--PO-blue)
      • [pdf - wang.github.io/high-entropy-minority-tokens-rlvr/)], 2025.06. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Training_with_High--Entropy_Tokens-green)
      • [pdf - Group/EPiC)], 2025.06. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/EPiC-blue) ![](https://img.shields.io/badge/Step_Shortcut-green)
      • [pdf - orange) ![](https://img.shields.io/badge/s1-blue)
      • [pdf - sg/understand-r1-zero)], 2025.03. ![](https://img.shields.io/badge/COLM2025-orange) ![](https://img.shields.io/badge/Dr.GRPO-blue)
      • [pdf - R1)], 2025.03. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Light--R1-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/FastCuRL-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/TBA-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/CPPO-blue)
      • [pdf - NLP/LIMO)], 2025.02. ![](https://img.shields.io/badge/COLM2025-orange) ![](https://img.shields.io/badge/LIMO-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/SAR-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/CDE-blue)
      • [pdf - SIA/DAPO)], [[homepage](https://dapo-sia.github.io/)], 2025.03. ![](https://img.shields.io/badge/NeurIPS2025-orange) ![](https://img.shields.io/badge/DAPO-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/AReal-blue)
      • [pdf - ustc/Alpha-RL)], 2025.10. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/AlphaRL-blue)
    • Latent Chain-of-Thought

      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/PCCoT-blue)
      • [pdf - ai-lab/Soft-Thinking)], 2025.05. ![](https://img.shields.io/badge/NeurIPS2025-orange) ![](https://img.shields.io/badge/Soft--Thinking-blue)
      • [pdf - NLP/Awesome-Latent-CoT)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Survey-green)
      • [pdf - latent-reasoning.github.io/)], 2025.05. ![](https://img.shields.io/badge/NeurIPS2025-orange) ![](https://img.shields.io/badge/CoLaR-blue)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/PHD--Transformer-blue)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Filler-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/ImplicitCoT-blue)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Analysis-green)
      • [pdf - Memory-and-Reasoning)], 2024.11. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/CCoT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Heima-blue) ![](https://img.shields.io/badge/Multimodal-green)
      • [pdf - orange) ![](https://img.shields.io/badge/ITT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/LightThinker-blue)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/CODI-blue)
      • [pdf - rg/recurrent-pretraining)], 2025.02. ![](https://img.shields.io/badge/NeurIPS2025-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/CoCoMix-blue) ![](https://img.shields.io/badge/Pretrain-green)
      • [pdf - orange) ![](https://img.shields.io/badge/LTM-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/RELAY-blue)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/SoftCoT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Analysis-green)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/ReaRec-blue) ![](https://img.shields.io/badge/Sequential_Recommendation-green)
      • [pdf - orange) ![](https://img.shields.io/badge/MCOUT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/COCONUT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/HRPO-blue)
      • [pdf - orange)
      • [pdf - CoT)], 2025.09. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/SIM--CoT-blue)
    • Small & Large Reasoning Model Collaboration

      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Hawkeye-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/SMART-blue)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/FoReaL--Decoding-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/ProxyThinker-blue) ![](https://img.shields.io/badge/CoTs_from_Multimodal_SRMs-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Thought_Manipulation-blue) ![](https://img.shields.io/badge/CoTs_from_SRMs-green)
      • [pdf - orange) ![](https://img.shields.io/badge/SplitReason-blue) ![](https://img.shields.io/badge/offloading_challenging_CoT_parts_to_LRMs-green)
    • Optimal Test-Time Scaling

      • [pdf - orange) ![](https://img.shields.io/badge/DORA-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/SPECS-blue)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/BEST--Route-blue)
      • [pdf - gh98/Guided-by-Gut)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - guided-search)], 2025.05. ![](https://img.shields.io/badge/NeurIPS2025-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Shortest_N-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Shortest_N-green)
      • [pdf - First-Search)], 2025.06. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/LFS-blue)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - wyz/inference_scaling)], [[homepage](https://thu-wyz.github.io/inference-scaling/)], 2024.08. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Over--thinking-green)
      • [pdf - time-scaling-eval)], 2025.02. ![](https://img.shields.io/badge/ACL2025-orange) ![](https://img.shields.io/badge/Over--thinking-green)
      • [pdf - orange) ![](https://img.shields.io/badge/InferenceTimePessimism-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Survey-green)
      • [pdf - genrm-scaling)], 2025.04. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Z1-blue)
      • [pdf - AIRe/MRT)], [[homepage](https://cohenqu.github.io/mrt.github.io/)], 2025.03. ![](https://img.shields.io/badge/ICML2025-orange) ![](https://img.shields.io/badge/MRT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/DeepConf-blue)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/ParaThinker-blue)
    • Parallel Thinking

      • [pdf - AI-Lab/Multiverse)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/SPRINT-blue) ![](https://img.shields.io/badge/Parallel_Execution-green)
      • [pdf - Reasoning/APR)], 2025.04. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/APR-blue) ![](https://img.shields.io/badge/Parallel_Computation-green)
      • [pdf - orange)
      • [pdf - decoding-in-one-sequence)], 2024.03. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Parallel_Decoding-green)
      • [pdf - research/sot)], [[homepage](https://sites.google.com/view/sot-llm)], 2023.06. ![](https://img.shields.io/badge/ICLR2024-orange) ![](https://img.shields.io/badge/SoT-blue)
      • [pdf - orange)
      • [pdf - orange)
    • Reasoning Shortcuts

      • [pdf - orange) ![](https://img.shields.io/badge/TokenSkip-blue) ![](https://img.shields.io/badge/Token_Shortcut-green)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/Break_the_Chain-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Step_Shortcut-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Step_Shortcut-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Token_Shortcut-green)
      • [pdf - orange) ![](https://img.shields.io/badge/DRP-blue) ![](https://img.shields.io/badge/Step_Shortcut-green)
      • [pdf - yibo/R1-Compress)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/R1--Compress-blue) ![](https://img.shields.io/badge/Step_Shortcut-green)
      • [pdf - All-Thinking-Tokens)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Token_Shortcut-green)
      • [pdf - NLP/LIMOPro)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/LIMOPro-blue) ![](https://img.shields.io/badge/Step_Shortcut-green)
      • [pdf - orange) ![](https://img.shields.io/badge/DTO-blue) ![](https://img.shields.io/badge/Step_Shortcut-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Step_Shortcut-green)
    • Benchmarks

      • [pdf - orange) ![](https://img.shields.io/badge/S1--Bench-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/LLMThinkBench-blue)
      • [pdf - bench.github.io/)], [[code](https://github.com/ZhiyuanLi218/Think-Bench)], 2025.04. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/THINK--Bench-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/THOUGHTTERMINATOR-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/DNA_Bench-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/OptimalThinkingBench-blue)
    • Multimodal Reasoning Efficiency

      • [pdf - 4B)], [[huggingface](https://huggingface.co/YannQi/R-4B)], 2025.08. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/R--4B-blue)
      • [pdf - zju/PixelThink)], [[homepage](https://pixelthink.github.io/)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/PixelThink-blue)
      • [pdf - orange)
      • [pdf - fuxi.github.io/projects/uni-cot/)], 2025.09. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Uni--cot-blue)
    • Adaptive Thinking

      • [pdf - orange) ![](https://img.shields.io/badge/SynapseRoute-blue)
      • [pdf - orange)
      • [pdf - KEG/AdaptThink)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/AdaptThink-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Thinkless-blue)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/ThinkSwitcher-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/CAR-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/ASRR-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/AdaCoT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/AutoL2S-blue)
      • [pdf - project.github.io/)], [[code](https://github.com/ASTRAL-Group/AlphaOne)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/AlphaOne-blue)
      • [pdf - Lab/OThink-R1)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/OThink--R1-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/SwitchCoT-blue)
      • [pdf - fib-lab/Token_Signature)], 2025.06. ![](https://img.shields.io/badge/ICML2025-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/SelfThink-blue)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - arm.github.io/arm/)], [[code](https://github.com/TEAM-ARM/ARM)], 2025.05. ![](https://img.shields.io/badge/NeurIPS2025-orange) ![](https://img.shields.io/badge/ARM-blue)
    • Sparse Attention & KV Cache

      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/R--KV-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/SeerAttention--R-blue)
      • [pdf - orange)
    • Efficient Sampling

      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - Labs/SpeculativeRejection)], 2024.10. ![](https://img.shields.io/badge/NIPS2024-orange) ![](https://img.shields.io/badge/Speculative_Rejection-blue) ![](https://img.shields.io/badge/Best--of--N-green)
      • [pdf - github-00/LLM-Predictive-Decoding)], 2024.10. ![](https://img.shields.io/badge/ICLR2025-orange) ![](https://img.shields.io/badge/Predictive--Decoding-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/FastMCTS-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/DPTS-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/FETCH-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/ST--BoN-blue) ![](https://img.shields.io/badge/Best--of--N-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Self--taught_Lookahead-blue)
      • [pdf - Decoding)], 2025.03. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/ϕ--Decoding-blue)
    • Reasoning Step Decomposition

      • [pdf - yw/Markov-Chain-of-Thought)], 2024.10. ![](https://img.shields.io/badge/NAACL2025-orange) ![](https://img.shields.io/badge/MCoT-blue) ![](https://img.shields.io/badge/Intermediate_CoT_Summarization-green)
      • [pdf - orange) ![](https://img.shields.io/badge/AoT-blue)
      • [pdf - -FM_workshop-orange) ![](https://img.shields.io/badge/DISC-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/SCoT-blue) ![](https://img.shields.io/badge/Multimodal-green)
      • [pdf - orange) ![](https://img.shields.io/badge/AR-blue)
    • Efficient Self-Consistency

      • [pdf - orange) ![](https://img.shields.io/badge/ESC-blue)
      • [pdf - -Findings-orange) ![](https://img.shields.io/badge/DSC-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Path--Consistency-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/RASC-blue)
      • [pdf - Huang/Self-Calibration)], 2025.02. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Self--Calibration-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/RPC-blue)
      • [pdf - -findings-orange) ![](https://img.shields.io/badge/CISC-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Slim--SC-blue)
    • Long-Context Reasoning Efficiency

      • [pdf - orange) ![](https://img.shields.io/badge/OmniKV-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/InftyThink-blue) ![](https://img.shields.io/badge/Intermediate_CoT_Summarization-green)
    • Applications

      • [pdf - orange) ![](https://img.shields.io/badge/Audio-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Code-green)
      • [pdf - orange) ![](https://img.shields.io/badge/Tool_Use-green)
  • Resources

  • Keywords Convention

  • Blog & Project

    • Applications

  • Talks