面向金融推理的依赖感知思维链压缩
Dependency-Aware Chain-of-Thought Compression for Financial Reasoning
- University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
- Syracuse University（雪城大学）
- Northeastern University（东北大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对思维链在金融推理中成本高的问题，提出HSDN分层语义蒸馏网络压缩推理链，在AFAC2025基准上准确率91.0%、压缩率68.4%，优于基线，验证了图引导压缩的有效性。
AI中文摘要：
思维链提示可提升复杂推理能力，但其冗长的中间轨迹会产生大量推理成本，阻碍在金融场景的实际部署。本文提出分层语义蒸馏网络（Hierarchical Semantic Distillation Network，HSDN），用于压缩推理链，同时保持答案准确性和逻辑连贯性。该框架结合语义分割、依赖图构建、双编码器重要性评分、受约束的片段选择及局部边界重写，仅使用冻结的Qwen3 4B模型进行特征提取和最终答案生成，压缩过程保持结构化且可解释。在AFAC2025基准上，HSDN实现91.0%的准确率和68.4%的压缩率，在综合得分和推理连贯性上优于强大的压缩基线。结果表明，图引导压缩对高风险金融推理任务有效。
英文摘要：
Chain of thought prompting improves complex reasoning, but its long intermediate traces create substantial inference cost and hinder practical deployment in financial settings. We present a Hierarchical Semantic Distillation Network, HSDN, for compressing reasoning chains while preserving answer accuracy and logical coherence. The framework combines semantic segmentation, dependency graph construction, dual encoder importance scoring, constrained segment selection, and local boundary rewriting. A frozen Qwen3 4B model is used only for feature extraction and final answer generation, while the compression process remains structured and interpretable. On the AFAC2025 benchmark, HSDN achieves 91.0% accuracy with 68.4% compression, outperforming strong compression baselines in overall score and reasoning coherence. The results show that graph guided compression is effective for high stakes financial reasoning tasks.
