mlsanityData to 14 Oct 2026

AutoL2S: Auto Long-Short Reasoning for Efficient Large Language Models

arXiv 2505.22662 ↗First posted 28 May 2025cs.CL cs.LGcited by 12

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

Reasoning-capable large language models (LLMs) achieve strong performance on complex tasks but often exhibit overthinking after distillation, generating unnecessarily long chain-of-thought (CoT) reasoning even for simple inputs and incurring high inference cost. However, naively shortening reasoning length can degrade reasoning accuracy, as concise reasoning may be insufficient for certain inputs and lacks explicit supervision. We propose Auto Long-Short Reasoning (AutoL2S), a distillation framework that empowers non-reasoning LLMs to think thoroughly but only when necessary. AutoL2S first learns a lightweight switching token with verified long-short CoTs to enable instance-wise long-short reasoning selection. Then it leverages long-short reasoning rollouts induced by a switching token in a GRPO-style loss to improve reasoning efficiency while maintaining accuracy. Experiments demonstrate that AutoL2S effectively reduces reasoning length up to 71% with minimal accuracy loss, yielding markedly better trade-off in token length and inference time while preserving accuracy.

Key facts. Cited by 12 papers in the map (data through 14 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 28 of its 38 references that are in the map. First posted on arXiv 28 May 2025.

Builds on

28 of 38 references.

Closest ancestors · 13

Same direction, newest first

  1. 2025Thinkless: LLM Learns When to Think19 May
  2. 2025ThinkPrune: Pruning Long Chain-of-Thought of LLMs via Reinforcement Learning2 Apr
  3. 2025Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models20 Mar
  4. 2025L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning6 Mar
  5. 2025DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Models6 Mar
  6. 2025Thinking Preference Optimization17 Feb
  7. 2025TokenSkip: Controllable Chain-of-Thought Compression in LLMs17 Feb
  8. 2025CoT-Valve: Length-Compressible Chain-of-Thought Tuning13 Feb
  9. 2025Demystifying Long Chain-of-Thought Reasoning in LLMs5 Feb
  10. 2025s1: Simple test-time scaling31 Jan
  11. 2025O1-Pruner: Length-Harmonizing Fine-Tuning for O1-Like Reasoning Pruning22 Jan
  12. 2024Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs30 Dec
1 more
  1. 2024Token-Budget-Aware LLM Reasoning24 Dec

Other references · 15

  1. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  2. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  3. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  4. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  5. 2024Qwen2.5 Technical Reportcited by 4,942
  6. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,848
  7. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,210
  8. 2024OlympiadBench: A Challenging Benchmark for Promoting AGI with Olympiad-Level Bilingual Multimodal Scientific Problemscited by 1,136
  9. 2025LIMO: Less is More for Reasoningcited by 498
  10. 2025The Lessons of Developing Process Reward Models in Mathematical Reasoningcited by 415
  11. 2023LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Modelscited by 287
  12. 2024Distilling System 2 into System 1cited by 159
3 more
  1. 2024Can Language Models Learn to Skip Steps?cited by 59
  2. 2025Stepwise Perplexity-Guided Refinement for Efficient Chain-of-Thought Reasoning in Large Language Modelscited by 51
  3. 2025Z1: Efficient Test-time Scaling with Codecited by 35

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 7; 2026: 5.

12 papers.

  1. 2025A Survey of Context Engineering for Large Language Modelscited by 126
  2. 2025Don't Overthink It: A Survey of Efficient R1-style Large Reasoning Modelscited by 37
  3. 2025Aware First, Think Less: Dynamic Boundary Self-Awareness Drives Extreme Reasoning Efficiency in Large Language Modelscited by 13
  4. 2026Neural Chain-of-Thought Search: Searching the Optimal Reasoning Path to Enhance Large Language Modelscited by 8
  5. 2026A Survey on Evaluating Quality and Trustworthiness in LLM-Generated Datacited by 6
  6. 2025PEAR: Phase Entropy Aware Reward for Efficient Reasoningcited by 5
  7. 2025DART: Difficulty-Adaptive Reasoning Truncation for Efficient Large Language Modelscited by 5
  8. 2025DRQA: Dynamic Reasoning Quota Allocation for Controlling Overthinking in Reasoning Large Language Modelscited by 3
  9. 2025Fast, Slow, and Tool-augmented Thinking for LLMs: A Reviewcited by 2
  10. 2026HyPER: Bridging Exploration and Exploitation for Scalable LLM Reasoning with Hypothesis Path Expansion and Reductioncited by 0
  11. 2026Breaking the Reward Barrier: Accelerating Tree-of-Thought Reasoning via Speculative Explorationcited by 0
  12. 2026Efficient Reasoning via Constrained Optimization in Latent Spacecited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Efficient Inference for Large Reasoning Models: A Survey
  2. 2026A State-Transition Framework for Efficient LLM Reasoning
  3. 2026Stabilizing Efficient Reasoning with Step-Level Advantage Selection
  4. 2025Advancing Reasoning in Large Language Models: Promising Methods and Approaches
  5. 2025Don't Overthink it. Preferring Shorter Thinking Chains for Improved LLM Reasoning
  6. 2025Prolonged Reasoning Is Not All You Need: Certainty-Based Adaptive Routing for Efficient LLM/MLLM Reasoning
  7. 2026The Art of Efficient Reasoning: Data, Reward, and Optimization
  8. 2025Deconstructing Long Chain-of-Thought: A Structured Reasoning Optimization Framework for Long CoT Distillation