通过思想相似性剪枝，让自一致性推理更快更省显存。
Slim-SC: Thought Pruning for Efficient Scaling with Self-Consistency
- 按思维链相似度逐轮剔除冗余推理路径
- 在3个理科数据集上降低45%延迟、26%显存占用
- 适合需要高效推理的部署场景
近期，测试时扩展（TTS）因其无需重训练即可提升大模型推理性能而受到关注。其中自一致性（SC）通过并行生成多条推理链并以多数投票决定答案，效果显著，但计算开销巨大。现有加速方法多依赖模型置信度或启发式规则，缺乏实证支持。本文首次从理论和实验层面分析了SC的效率瓶颈，并提出Slim-SC：一种基于思维链间相似性的分步剪枝策略，可有效识别并移除冗余推理路径。在三个STEM推理数据集及两种最新LLM架构上的实验表明，使用R1-Distill时，Slim-SC可将推理延迟降低最多45%，显存占用减少26%，同时保持或提升准确率，为自一致性提供了一种简单高效的替代方案。
原文摘要 · Abstract (English)
Recently, Test-Time Scaling (TTS) has gained increasing attention for improving LLM reasoning performance at test time without retraining the model. A notable TTS technique is Self-Consistency (SC), which generates multiple reasoning chains in parallel and selects the final answer via majority voting. While effective, the order-of-magnitude computational overhead limits its broad deployment. Prior attempts to accelerate SC mainly rely on model-based confidence scores or heuristics with limited empirical support. For the first time, we theoretically and empirically analyze the inefficiencies of SC and reveal actionable opportunities for improvement. Building on these insights, we propose Slim-SC, a step-wise pruning strategy that identifies and removes redundant chains using inter-chain similarity at the thought level. Experiments on three STEM reasoning datasets and two recent LLM architectures show that Slim-SC reduces inference latency and KVC usage by up to 45% and 26%, respectively, with R1-Distill, while maintaining or improving accuracy, thus offering a simple yet efficient TTS alternative for SC.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。