LEAD: Length-Efficient Adaptive and Dynamic Reasoning for Large Language Models
On the map: LLM reasoning → Test-time compute and efficient thinking
Abstract (arXiv)
Large reasoning models, such as OpenAI o1 and DeepSeek-R1, tend to become increasingly verbose as their reasoning capabilities improve. These inflated Chain-of-Thought (CoT) trajectories often exceed what the underlying problems require, wasting compute, latency, and context budgets. While introducing length-based efficiency rewards during reinforcement learning offers a natural remedy, existing methods struggle with two fundamental challenges: the optimal balance between correctness and efficiency is non-stationary throughout training, and intrinsic reasoning budgets vary drastically across problems. Relying on static reward weights and global length constraints inevitably forces a compromise between degraded accuracy and unrealized compression. To overcome these limitations, we propose LEAD (Length-Efficient Adaptive and Dynamic reasoning), a method that replaces static heuristics with online, self-adaptive mechanisms. LEAD dynamically calibrates the correctness-efficiency trade-off at each step using a Potential-Scaled Instability, directing optimization capacity to the most informative learning signal. Furthermore, it estimates an adaptive per-problem target length online based on the model's own correct rollouts, applying a symmetric efficiency reward that penalizes both overthinking and over-compression. Evaluated on five mathematical reasoning benchmarks, LEAD achieves the highest accuracy and Accuracy-Efficiency Score among RL-trained efficient-reasoning methods while producing substantially shorter outputs than the base model.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 30 of its 40 references that are in the map. First posted on arXiv 10 May 2026.
Builds on
30 of 40 references.
Closest ancestors · 12
Same direction, newest first
- 2025The Art of Scaling Test-Time Compute for Large Language Models1 Dec
- 2025e1: Learning Adaptive Control of Reasoning Effort30 Oct
- 2025SmartThinker: Learning to Compress and Preserve Reasoning by Step-Level Length Control6 Jul
- 2025Just Enough Thinking: Efficient Reasoning with Adaptive Length Penalties Reinforcement Learning5 Jun
- 2025Plan and Budget: Effective and Efficient Test-Time Scaling on Reasoning Large Language Models22 May
- 2025Learn to Reason Efficiently with Adaptive Length-based Reward Shaping21 May
- 2025SelfBudgeter: Adaptive Token Allocation for Efficient LLM Reasoning16 May
- 2025ShorterBetter: Guiding Reasoning Models to Find Optimal Inference Length for Efficient Reasoning30 Apr
- 2025ThinkPrune: Pruning Long Chain-of-Thought of LLMs via Reinforcement Learning2 Apr
- 2025L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning6 Mar
- 2025O1-Pruner: Length-Harmonizing Fine-Tuning for O1-Like Reasoning Pruning22 Jan
- 2024Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs30 Dec
Other references · 18
- 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
- 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
- 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
- 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
- 2024Qwen2.5 Technical Reportcited by 4,942
- 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
- 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 2,057
- 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 1,407
- 2024OpenAI o1 System Cardcited by 1,244
- 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,210
- 2025Training Language Models to Reason Efficientlycited by 247
- 2026GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimizationcited by 193
6 more
- 2025A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibilitycited by 93
- 2025Sample More to Think Less: Group Filtered Policy Optimization for Concise Reasoningcited by 75
- 2025Learning to Optimize Multi-Objective Alignment Through Dynamic Reward Weightingcited by 23
- 2025HAPO: Training Language Models to Reason Concisely via History-Aware Policy Optimizationcited by 19
- 2025DRPO: Efficient Reasoning via Decoupled Reward Policy Optimizationcited by 16
- 2025TokenFlow: Responsive LLM Text Streaming Serving under Request Burst via Preemptive Schedulingcited by 10
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026The Art of Efficient Reasoning: Data, Reward, and Optimization
- 2026Correct, Concise and Complete: Multi-stage Training For Adaptive Reasoning
- 2025Thinking Fast and Right: Balancing Accuracy and Reasoning Length with Adaptive Rewards
- 2025AALC: Large Language Model Efficient Reasoning via Adaptive Accuracy-Length Control
- 2026SmartThinker: Progressive Chain-of-Thought Length Calibration for Efficient Large Language Model Reasoning
- 2025Leash: Adaptive Length Penalty and Reward Shaping for Efficient Large Reasoning Model
- 2025Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models
- 2026Stabilizing Efficient Reasoning with Step-Level Advantage Selection