根据模型自信度动态切换推理，提升大模型效率与准确率
Prolonged Reasoning Is Not All You Need: Certainty-Based Adaptive Routing for Efficient LLM/MLLM Reasoning
- 依据困惑度判断是否启动长链条推理，避免无效思考
- 在多模态问答和文本推理任务中，精度高于纯短答或纯长答
- 适合追求高效推理的部署场景，尤其对简单任务有明显加速
近期推理技术显著提升了大语言模型（LLMs）和多模态大语言模型（MLLMs）在各类任务中的表现。然而，过度依赖思维链（CoT）推理会降低模型性能并导致输出过长，影响效率。本文发现，延长推理并不总能提升准确率，甚至在简单任务上会导致性能下降。为此，我们提出基于置信度的自适应推理框架（CAR），根据模型困惑度动态切换短答案与长推理模式：先生成短答案并评估其困惑度，仅当模型低自信（即高困惑度）时才触发推理。在多个多模态视觉问答（VQA）和关键信息提取（KIE）基准及文本推理数据集上的实验表明，CAR在准确率与效率之间取得最优平衡，优于纯短答或纯长答方法。
原文摘要 · Abstract (English)
Recent advancements in reasoning have significantly enhanced the capabilities of Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs) across diverse tasks. However, excessive reliance on chain-of-thought (CoT) reasoning can impair model performance and brings unnecessarily lengthened outputs, reducing efficiency. Our work reveals that prolonged reasoning does not universally improve accuracy and even degrade performance on simpler tasks. To address this, we propose Certainty-based Adaptive Reasoning (CAR), a novel framework that dynamically switches between short answers and long-form reasoning based on the model perplexity. CAR first generates a short answer and evaluates its perplexity, triggering reasoning only when the model exhibits low confidence (i.e., high perplexity). Experiments across diverse multimodal VQA/KIE benchmarks and text reasoning datasets show that CAR outperforms both short-answer and long-form reasoning approaches, striking an optimal balance between accuracy and efficiency.
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。