REVIEW 42 cited by
O1-Pruner: Length-Harmonizing Fine-Tuning for O1-Like Reasoning Pruning
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Recently, long-thought reasoning LLMs, such as OpenAI's O1, adopt extended reasoning processes similar to how humans ponder over complex problems. This reasoning paradigm significantly enhances the model's problem-solving abilities and has achieved promising results. However, long-thought reasoning process leads to a substantial increase in inference time. A pressing challenge is reducing the inference overhead of long-thought LLMs while ensuring accuracy. In this paper, we experimentally demonstrate that long-thought reasoning models struggle to effectively allocate token budgets based on problem difficulty and reasoning redundancies. To address this, we propose Length-Harmonizing Fine-Tuning (O1-Pruner), aiming at minimizing reasoning overhead while maintaining accuracy. This effective fine-tuning method first estimates the LLM's baseline performance through pre-sampling and then uses RL-style fine-tuning to encourage the model to generate shorter reasoning processes under accuracy constraints. This allows the model to achieve efficient reasoning with lower redundancy while maintaining accuracy. Experiments on various mathematical reasoning benchmarks show that O1-Pruner not only significantly reduces inference overhead but also achieves higher accuracy, providing a novel and promising solution to this challenge. Our code is coming soon at https://github.com/StarDewXXX/O1-Pruner
Forward citations
Cited by 42 Pith papers
-
Is Long-to-Short a Free Lunch? Investigating Inconsistency and Reasoning Efficiency in LRMs
Efficient reasoning methods that skip or compress a model's thinking increase behavioral inconsistency in large reasoning models, according to a new benchmark called ICBENCH.
-
EvoThink: Evolving Thinking in Large Reasoning Models via Self-Pruning and Aha-Moment Preference Optimization
EvoThink reduces LRM overthinking by pruning redundant atomic reasoning steps (SPT) and training on diversity-selected wrong-to-right mutation data (AMPO), cutting tokens and improving math/code accuracy.
-
OS-Pruner: Pruning Chains-of-Thought of Reasoning Models via Optimal Stopping
A lightweight optimal-stopping policy on frozen reasoning LLMs cuts CoT length 20–60% with minimal accuracy loss by trading answer correctness against token cost via a tunable λ.
-
ConPress: Learning Efficient Reasoning from Multi-Question Contextual Pressure
Multi-question prompts elicit shorter chain-of-thought traces, and fine-tuning on those traces transfers the compression to single-question reasoning.
-
Knowledge Graph Augmented Large Language Models for Disease Prediction
A knowledge-graph-guided chain-of-thought fine-tuning pipeline for EHR disease prediction improves accuracy on transfer and clinician-rated explanation quality, but only modestly improves ranking metrics.
-
Understanding the Effects of Distractors on Reasoning Vision-Language Models
Adding visual distractors to images lowers reasoning-VLM accuracy while leaving reasoning length essentially flat, with the accuracy loss traced to the share of reasoning-trace attributes devoted to distractors.
-
Probing the Difficulty Perception Mechanism of Large Language Models
LLMs linearly encode math-problem difficulty in their final-token representations, and specific final-layer attention heads are specialized for easy vs hard problems.
-
Reconsidering Overthinking: Penalizing Internal and External Redundancy in CoT Reasoning
A dual-penalty RL method that compresses chain-of-thought traces by separately penalizing internal semantic stagnation and external post-answer continuation reduces reasoning length by about 40% while preserving accur...
-
LAPO: Internalizing Reasoning Efficiency via Length-Adaptive Policy Optimization
LAPO is a two-stage RL method that reduces reasoning token usage by up to 40.9% while slightly improving accuracy, by learning per-problem length targets from the model's own correct answers and prompting the model to...
-
Do Thinking Tokens Help or Trap? Towards More Efficient Large Reasoning Model
Suppressing 'thinking tokens' in a 1.5B reasoning model preserves accuracy while cutting tokens, and the proposed DuP-PO RL method improves both accuracy and efficiency over GRPO.
-
AALC: Large Language Model Efficient Reasoning via Adaptive Accuracy-Length Control
AALC, an accuracy-aware length reward, reduces reasoning-model response length by over 50% on math benchmarks while preserving or improving accuracy.
-
How Far Are We from Optimal Reasoning Efficiency?
The authors define a reasoning efficiency frontier and a gap metric (REG), then train models with REO-RL to shrink the gap by at least 50% with only small accuracy losses.
-
EPiC: Towards Lossless Speedup for Reasoning Training through Edge-Preserving CoT Condensation
Keeping only the first and last parts of chain-of-thought training traces preserves math reasoning accuracy on MATH500 at half the sequence length, but not on harder benchmarks.
-
PixelThink: Towards Efficient Chain-of-Pixel Reasoning
A soft token-budget penalty driven by task difficulty and model confidence roughly halves the reasoning tokens used by segmentation MLLMs while slightly improving mask accuracy on ReasonSeg-derived benchmarks.
-
From Token to Action: State Machine Reasoning to Mitigate Overthinking in Information Retrieval
A state-machine framework that replaces token-level chain-of-thought with discrete query-refinement and reranking actions reduces token use by 74% while improving nDCG@10 on retrieval benchmarks.
-
Hume: Introducing System-2 Thinking in Visual-Language-Action Model
A dual-system vision-language-action model that improves robot control by ranking multiple sampled action chunks with a learned value function before fast execution.
-
Adaptive Deep Reasoning: Triggering Deep Thinking When Needed
A reinforcement learning recipe with adaptive group-wise rewards and a first-token margin loss lets a 7B math LLM autonomously choose between short and long chain-of-thought reasoning per problem.
-
System-1.5 Reasoning: Traversal in Language and Latent Spaces with Dynamic Shortcuts
System-1.5 Reasoning lets LLMs reason in latent space with early exits and step-skipping, matching chain-of-thought accuracy at over 20x speedup on GSM8K and StrategyQA.
-
Thinking Fast and Right: Balancing Accuracy and Reasoning Length with Adaptive Rewards
An adaptive length-penalty reward for RL-trained LLMs reduces reasoning length by over 50% with small accuracy loss by automatically tightening and relaxing the penalty based on the model's current accuracy.
-
VeriThinker: Learning to Verify Makes Reasoning Model Efficient
VeriThinker shows that fine-tuning a reasoning model only on a solution-verification task reduces chain-of-thought length on MATH500 and AIME by 20-45% while preserving or slightly improving accuracy.
-
ConciseRL: Conciseness-Guided Reinforcement Learning for Efficient Reasoning Models
Using an LLM judge to reward semantic conciseness during reinforcement learning makes 1.5B reasoning models produce much shorter traces with comparable or better accuracy.
-
R1-ShareVL: Incentivizing Reasoning Capability of Multimodal Large Language Models via Share-GRPO
Share-GRPO creates paraphrased and visually augmented versions of reasoning questions and shares answers and reward signals across versions, improving multimodal reasoning without cold-start SFT.
-
Is Your Model Thinking or Just Stagnating? PUMA: Diagnosing Reasoning Pathology via Phase-Momentum Alignment
PUMA detects when a reasoning model's entropy drop aligns with hidden-state momentum, truncates at that point, and reports improved accuracy-efficiency on 1.5B-32B reasoning models.
-
Masked Distillation: Internalizing the Chain-of-Thought in Language Models
Masked distillation shows a student LLM can internalize its teacher's chain-of-thought on GSM8K but not on Countdown; emitting the last 30% of the trace as a 'suffix scaffold' recovers teacher-level accuracy with fewe...
-
MixReasoning: Switching Modes to Think
A reasoning model can cut token usage by roughly a third to a half without losing accuracy if it switches to brief output on low-uncertainty steps and expands only at high-uncertainty forks.
-
ThinkDial: An Open Recipe for Controlling Reasoning Effort in Large Language Models
An open training recipe that teaches a 32B LLM to reason in three adjustable modes, cutting token use by half or three quarters with small accuracy losses.
-
Think in Blocks: Adaptive Reasoning from Direct Response to Deep Reasoning
A training framework that makes LLMs predict an integer reasoning budget as a number of blocks and adapt reasoning depth to task difficulty.
-
Failure Cases Are Better Learned But Boundary Says Sorry: Facilitating Smooth Perception Change for Accuracy-Robustness Trade-Off in Adversarial Training
A paper whose abstract describes new adversarial training experiments, but whose full text is a different paper on CoT compression, leaving the claims unsupported.
-
REST: Stress Testing Large Reasoning Models by Asking Multiple Problems at Once
Asking a reasoning model several problems at once reveals large accuracy drops and exposes differences that single-question benchmarks miss.
-
CoRE: Enhancing Metacognition with Label-free Self-evaluation in LRMs
A training-free and label-free detector of cyclic hidden-state patterns triggers early exit during chain-of-thought reasoning, reducing token length while mostly preserving or improving accuracy.
-
Enhancing Large Language Models through Structured Reasoning
Structured reasoning tags plus a max-flow reward let a 1.5B model match the math accuracy of models trained for far longer, but the gains are within statistical noise.
-
Efficient Reasoning Through Suppression of Self-Affirmation Reflections in Large Reasoning Models
Suppressing low-probability 'wait' tokens that trigger self-affirmation reflections shortens reasoning-model output by 8-18% with roughly unchanged accuracy.
-
Fast on the Easy, Deep on the Hard: Efficient Reasoning via Powered Length Penalty
Reward shaping with a powered length penalty makes LLMs answer easy questions with far fewer tokens while preserving or slightly improving accuracy on hard math benchmarks.
-
Don't Think Longer, Think Wisely: Optimizing Thinking Dynamics for Large Reasoning Models
DTO constructs compressed, ground-truth-curated reasoning trajectories from LRM outputs and uses them as preferred responses in SimPO, yielding up to 12% accuracy gains and roughly 40% token reduction on math benchmarks.
-
Can Past Experience Accelerate LLM Reasoning?
LLMs can cut reasoning compute by up to 56% on similar questions when memory of past correct solutions is combined with adaptive early stopping, though the benchmark is small and synthetic.
-
ThinkSwitcher: When to Think Hard, When to Think Fast
ThinkSwitcher trains a lightweight module to predict when a reasoner should use short versus long chain-of-thought, reducing inference tokens by 20-30% with only a small accuracy drop.
-
A Practical Two-Stage Recipe for Mathematical LLMs: Maximizing Accuracy with SFT and Efficiency with Reinforcement Learning
Ten-epoch SFT then GRPO raises AIME/MATH-500 accuracy while reducing solution length, achieving 8th place on AIMO's private set.
-
Wait, We Don't Need to "Wait"! Removing Thinking Tokens Improves Reasoning Efficiency
Suppressing "Wait"-like reflection tokens at decode time reduces reasoning token counts by 27-51% across five R1-style model families, with mixed accuracy effects.
-
Pangu Embedded: An Efficient Dual-system LLM Reasoner with Metacognition
Pangu Embedded, a 7B reasoner trained with iterative distillation, RL, and an adaptive fast/slow thinking scheme, reports superior benchmark scores to similarly sized Qwen3-8B and GLM-4-9B.
-
FlashThink: An Early Exit Method For Efficient Reasoning
A verifier LLM decides when a reasoning model has thought enough, cutting reasoning tokens by about 77% on two large models with roughly unchanged accuracy across four benchmarks.
-
Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey
A comprehensive review that categorizes methods for shortening and adaptively triggering chain-of-thought reasoning in large language models.
- Optimizing Length Compression in Large Reasoning Models
Discussion (0). Sign in to comment.