SYNLAT：面向思维链推理的语法对齐文本-潜变量压缩
SYNLAT: Syntax-Aligned Text-Latent Compression for Chain-of-Thought Reasoning
浏览论文内容
中文总结 AI 辅助
SynLat通过语法对齐单元（SAUs）对齐压缩边界，在受限预算下保留答案关键信息，在12个任务组中达到或超过最强基线，并在强压缩下显著提升性能。
中文摘要 AI 辅助
长思维链（CoT）轨迹会带来巨大的输出令牌成本。在受限预算下，压缩必须保留答案关键信息，这使得边界放置成为核心问题。令牌级和固定长度边界可能分割连贯片段，如短语、公式和局部推导，而步骤级边界可能绑定需要不同压缩操作的内容。我们提出SynLat，一种文本-潜变量CoT框架，通过非重叠的语法对齐单元（SAUs）将压缩边界与句法结构对齐。一个答案条件的教师模型为单个压缩条件的学生模型构建渐进式KEEP/LATENT目标，该学生模型在推理时仅根据问题和请求的压缩级别生成混合推理。在两种Qwen3学生规模、标准CoT和长CoT组以及三个压缩级别下，SynLat在所有12个任务组聚合中达到或超过最强评估基线，并在报告的achieved-CR选择协议下在11个中严格领先。总体增益在MEDIUM压缩级别达到3.6/2.6个百分点，在HIGH压缩级别达到7.0/5.5个百分点（Qwen3-8B/14B），在更强压缩下优势更大，尤其在长CoT组中。
英文摘要
Long chain-of-thought (CoT) traces impose substantial output-token costs. Under constrained budgets, compression must preserve answer-critical information, making boundary placement central. Token-level and fixed-length boundaries can fragment coherent spans such as phrases, formulas, and local derivations, whereas step-level boundaries can bind content requiring different compression actions. We introduce SynLat, a text-latent CoT framework that aligns compression boundaries with syntactic structure through non-overlapping Syntax-Aligned Units (SAUs). An answer-conditioned Teacher constructs progressive KEEP/LATENT targets for a single compression-conditioned Student, which generates mixed reasoning from only the question and requested compression level at inference. Across two Qwen3 Student scales, Standard-CoT and Long-CoT groups, and three compression levels, SynLat matches or exceeds the strongest evaluated baseline in all 12 task-group aggregates and strictly leads in 11 under the reported achieved-CR selection protocol. Overall gains reach 3.6/2.6 points at MEDIUM and 7.0/5.5 points at HIGH for Qwen3-8B/14B, with larger advantages under stronger compression, particularly on Long-CoT groups.
发表机构
- Nanyang Technological University（南洋理工大学）
- Nankai University（南开大学）
机构由 AI 辅助整理，请以论文原文为准。
