paper-with-me

홈 › Papers

Reasoning Models Can be Accurately Pruned Via Chain-of-Thought Reconstruction

2025-09-15 · Ryan Lucas, Kayhan Behdin, Zhipeng Wang, Qingquan Song, Shao Tang, Rahul Mazumder arxiv

Reasoning language models such as DeepSeek-R1 produce long chain-of-thought traces during inference time which make them costly to deploy at scale. We show that using compression techniques such as neural network pruning produces greater performance loss than in typical language modeling tasks, and in some cases can make the model slower since they cause the model to produce more thinking tokens but with worse performance. We show that this is partly due to the fact that standard LLM pruning methods often focus on input reconstruction, whereas reasoning is a decode-dominated task. We introduce a simple, drop-in fix: during pruning we jointly reconstruct activations from the input and the model's on-policy chain-of-thought traces. This "Reasoning-Aware Compression" (RAC) integrates seamlessly into existing pruning workflows such as SparseGPT, and boosts their performance significantly. Code reproducing the results in the paper can be found at: https://github.com/RyanLucas3/RAC

📄 PDF Abstract BibTeX arXiv:2509.12464

Code (0)

등록된 구현이 없습니다.

Tasks

Network Pruning

Similar Papers 제목 키워드 기반

Cognitive Loop of Thought: Reversible Hierarchical Markov Chain for Efficient Mathematical Reasoning

2026-04-08 · Jia-Chen Zhang, Yu-Jie Xiong, Zheng Zhou arxiv

Multi-step Chain-of-Thought (CoT) has significantly advanced the mathematical reasoning capabilities of LLMs by leveraging explicit reasoning steps. However, the widespread adoption of Long CoT often results in sequence …

Mathematical Reasoning

DRP: Distilled Reasoning Pruning with Skill-aware Step Decomposition for Efficient Large Reasoning Models

2025-05-20 · YuXuan Jiang, Dawei Li, Frank Ferraro

While Large Reasoning Models (LRMs) have demonstrated success in complex reasoning tasks through long chain-of-thought (CoT) reasoning, their inference often involves excessively verbose reasoning traces, resulting in su…

GSM8KMathematical ReasoningToken ReductionTransfer Learning

Structured Thoughts For Improved Reasoning And Context Pruning

2026-07-11 · Zain Sarwar, Supriyo Chakraborty, Berkcan Kapusuzoglu, Chia-Hsuan Lee 외 arxiv

Large language models (LLMs) excel at generating long chains of thought, but long reasoning traces are often verbose and memory-inefficient. In this work, we introduce Structured Thoughts, a framework that organizes reas…

Making Slow Thinking Faster: Compressing LLM Chain-of-Thought via Step Entropy

2025-08-05 · Zeju Li, Jianyuan Zhong, Ziyang Zheng, Xiangyu Wen 외 arxiv

Large Language Models (LLMs) using Chain-of-Thought (CoT) prompting excel at complex reasoning but generate verbose thought processes with considerable redundancy, leading to increased inference costs and reduced efficie…

Reinforcement LearningMathematical Reasoning

Chain-of-Thought Compression Should Not Be Blind: V-Skip for Efficient Multimodal Reasoning via Dual-Path Anchoring

2026-01-20 · Dongxu Zhang, Yiding Sun, Cheng Tan, Wenbiao Yan 외 arxiv

While Chain-of-Thought (CoT) reasoning significantly enhances the performance of Multimodal Large Language Models (MLLMs), its autoregressive nature incurs prohibitive latency constraints. Current efforts to mitigate thi…

Multimodal Reasoning