Awesome-Efficient-Reasoning
Paper list for Efficient Reasoning.
https://github.com/hemingkx/Awesome-Efficient-Reasoning
Last synced: 15 days ago
JSON representation
-
Blog & Project
-
Keywords Convention
-
Papers
-
Adaptive Thinking
- [pdf - orange)
- [pdf - KEG/AdaptThink)], 2025.05.  
- [pdf - orange) 
- [pdf - orange)
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - project.github.io/)], [[code](https://github.com/ASTRAL-Group/AlphaOne)], 2025.05.  
- [pdf - Lab/OThink-R1)], 2025.05.  
- [pdf - orange) 
- [pdf - fib-lab/Token_Signature)], 2025.06. 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange)
- [pdf - arm.github.io/arm/)], [[code](https://github.com/TEAM-ARM/ARM)], 2025.05.  
- [pdf - orange)
- [pdf - orange)
- [pdf - orange)
- [pdf - orange)
- [pdf - orange)
- [pdf - orange)
- [pdf - orange)
- [pdf - orange)
-
Analysis
- [pdf - Impact-of-Reasoning-Step-Length-on-Large-Language-Models)], 2024.01.  
- [pdf - boundary)], 2024.10.  
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange)
- [pdf - 32B)], 2025.03. 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - lab/MiP-Overthinking)], 2025.04.  
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - USTC/LRM-plans-CoT)], 2025.06.  
- [pdf - orange) 
- [pdf - orange) 
- [pdf - ov-file)], 2025.06. 
- [pdf - latent-cot)], 2025.07. 
- [pdf - orange)
- [pdf - try-matters)], 2025.10. 
- [pdf - orange)
- [pdf - orange)
- [pdf - A-W/demystifying-hybrid-thinking)], 2025.10. 
-
Applications
-
Benchmarks
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - bench.github.io/)], [[code](https://github.com/ZhiyuanLi218/Think-Bench)], 2025.04.  
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
-
Efficient Sampling
- [pdf - Labs/SpeculativeRejection)], 2024.10.   
- [pdf - github-00/LLM-Predictive-Decoding)], 2024.10.  
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange)  
- [pdf - orange) 
- [pdf - Decoding)], 2025.03.  
- [pdf - orange)
- [pdf - orange)
- [pdf - orange)
- [pdf - orange)
-
Efficient Sampling Methods
- [pdf - orange)
-
Efficient Self-Consistency
- [pdf - orange) 
- [pdf - -Findings-orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - Huang/Self-Calibration)], 2025.02.  
- [pdf - -findings-orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - Probe)], 2026.02. 
- [pdf - orange)
- [pdf - System-AI-Lab/STEP)], 2026.01. 
-
Efficient Training
- [pdf - orange) 
- [pdf - NLP/LIMO)], 2025.02.  
- [pdf - R1)], 2025.03.  
- [pdf - SIA/DAPO)], [[homepage](https://dapo-sia.github.io/)], 2025.03.  
- [pdf - orange) 
- [pdf - sg/understand-r1-zero)], 2025.03.  
- [pdf - orange) 
- [pdf - orange) 
- [pdf - lime/verl)], 2025.04.  
- [pdf - orange) 
- [pdf - PO)], 2025.05.  
- [pdf - wang.github.io/high-entropy-minority-tokens-rlvr/)], 2025.06.  
- [pdf - Group/EPiC)], 2025.06.   
- [pdf - orange) 
- [pdf - orange) 
- [pdf - R1)], 2025.03.  
- [pdf - ai-lab.github.io/GRESO/)], [[code](https://github.com/Infini-AI-Lab/GRESO)], 2025.06.  
- [pdf - cpu/Question-Free-Fine-Tuning)], 2025.06.  
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - ustc/Alpha-RL)], 2025.10.  
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange)
- [pdf - han-lab/fastrl)], 2025.11.  
- [pdf - orange)
- [pdf - orange)
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange)
-
Latent Chain-of-Thought
- [pdf - orange)
- [pdf - orange)
- [pdf - orange)
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange)
- [pdf - orange) 
- [pdf - Memory-and-Reasoning)], 2024.11. 
- [pdf - orange) 
- [pdf - orange)  
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange)
- [pdf - orange) 
- [pdf - rg/recurrent-pretraining)], 2025.02. 
- [pdf - orange)  
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange)
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange)
- [pdf - orange)  
- [pdf - orange) 
- [pdf - NLP/Awesome-Latent-CoT)], 2025.05.  
- [pdf - ai-lab/Soft-Thinking)], 2025.05.  
- [pdf - latent-reasoning.github.io/)], 2025.05.  
- [pdf - orange)
- [pdf - orange)
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange)
- [pdf - orange) 
- [pdf - orange)
- [pdf - CoT)], 2025.09.  
- [pdf - orange) 
- [pdf - orange) 
- [pdf - orange)
- [pdf - nics/TaH)], 2025.11.  
- [pdf - GO-SOLO/Latent-SFT)], 2025.10. 
- [pdf - orange)
- [pdf - orange)
- [pdf - dmlr.github.io/)], [[code](https://github.com/eric-ai-lab/DMLR)], 2025.12. 
- [pdf - orange) 
- [pdf - orange)
- [pdf - LR)], 2025.10.   
- [pdf - Embodied-AGI/Mirage), [project](https://vlm-mirage.github.io/)], 2025.06.   
- [pdf - orange)  
- [pdf - GRPO-master)], 2025.11.  
- [pdf - Latent-master)], 2026.01.  
- [pdf - orange)  
- [pdf - huang.github.io/fast-thinkact/)], 2025.12.  
- [pdf - orange) 
- [pdf - orange) 
-
Long-Context Reasoning Efficiency
-
Long-to-Short Chain-of-Thought
- [pdf - of-symbol-planning)], 2023.05. 
- [pdf - concise-cot)], 2024.01.  
- [pdf - orange) 
- [pdf - -findings-orange)  
- [pdf - Valve)], 2025.02.  
- [pdf - reasoning)], 2025.02. 
- [pdf - of-draft)], 2025.02.   
- [pdf - CoT/compressed-cot)], 2025.03.  
- [pdf - orange)  
- [pdf - orange)  
- [pdf - to-Short-via-Model-Merging)], 2025.03.  
- [pdf - orange) 
- [pdf - Pruner)], 2025.01.  
- [pdf - orange) 
- [pdf - Labs/efficient-reasoning)], [[homepage](https://zanette-labs.github.io/efficient-reasoning/)], 2025.02. 
- [pdf - orange) 
-
Sub Categories
Long-to-Short Chain-of-Thought
133
Latent Chain-of-Thought
55
Efficient Training
36
Optimal Test-Time Scaling
27
Adaptive Thinking
25
Analysis
24
Small Reasoning Models & CoT Distillation
22
Applications
17
Multimodal Reasoning Efficiency
17
Survey
13
Speculative Decoding for CoT Efficiency
13
Other Work
12
Reasoning Shortcuts
12
Efficient Sampling
12
Efficient Self-Consistency
11
Parallel Thinking
11
Small & Large Reasoning Model Collaboration
10
Benchmarks
7
Sparse Attention & KV Cache
7
Reasoning Step Decomposition
5
Long-Context Reasoning Efficiency
5
Efficient Sampling Methods
1
Keywords
chain-of-thought
2
efficient-reasoning
2
reasoning
2
efficiency
1
large-language-models
1
large-reasoning-models
1
budget-aware
1
cot
1
efficient
1
long-cot
1
lrm
1
o1
1
o3
1
r1
1
slow-fast
1
compression
1
awesome-list
1
deep-learning
1
inference-time-compute
1
latent-representation
1
machine-learning
1
planning
1
recurrent-models
1
reinforcement-learning
1
transformers
1