Apr 2026· Proceedings of the 2026 3rd International Conference on Machine Learning and Intelligent Computing· 0 citations· 19 references
Computer Science
TL;DR
A Hierarchical Semantic Distillation Network, HSDN, is presented for compressing reasoning chains while preserving answer accuracy and logical coherence, and shows that graph guided compression is effective for high stakes financial reasoning tasks.
Abstract
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.
Long-horizon reasoning improves complex question answering but often requires expensive chain-of-thought generation and may transfer shortcut-sensitive fragments during distillation. This paper addresses the focused problem of learning a compact student model that preserves decision-relevant logic while reducing shortc...
Jiang-Bo Feng, Jin-Chao Guo, Yu-Xiang He et al.· International journal of pat...· 0 citations
This work introduces SynLat, a text-latent CoT framework that aligns compression boundaries with syntactic structure through non-overlapping Syntax-Aligned Units (SAUs) and generates mixed reasoning from only the question and requested compression level at inference.
Yi-Feng Zhao, Hong-Jun Yu, Shi-Bo Wang et al.· 0 citations
This work proposes ChainPrune, a novel reasoning path semantic structural optimization method to efficiently and controllably synthesize self-generated high-quality training data and incorporates a DPO-based preference learning method combined with supervised loss, effectively mitigating false reward suppression.
Wei-Hang Pan, Zhengxu Yu, Yuxiang Zhang et al.· 2 citations
Memory-Augmented Compression is proposed, a training-free framework that constructs reusable reasoning memories from historical traces and retrieves them as prefill-side scaffolds to compensate for information lost during compression.
Si-Meng Zhang, Yi-Long Chen, Wenyuan Zhang et al.· 0 citations
Differentiable Hierarchical Fusion is presented, a novel framework that merges reasoning models with efficient base models via differentiable optimization to produce concise, accurate outputs and introduces a dual-factor adaptive weighting mechanism to capture intra-block variance and inter-block importance hierarchies...
Chuan-Gen Gao, Wen-Lun Zhang, Shang-Kun Wang et al.· Proceedings of the Thirty-Fi...· 0 citations
TopoCompress is introduced, a training-free and model-agnostic framework that compresses long contexts by selecting coherent semantic spans by selecting coherent semantic spans and achieves performance comparable to the strongest baseline while using a 4x smaller compression budget.
