从幻觉到结构雪球效应：约束解码在LLM反思中的对齐税
From Hallucination to Structure Snowballing: The Alignment Tax of Constrained Decoding in LLM Reflection
AI总结：
研究探讨了在开放式推理任务中，约束解码通过强制结构反思是否能避免错误传播，发现反而引发结构雪球效应，并揭示了约束解码的对齐税问题。
摘要：
大型语言模型（LLM）的内在自我校正经常在开放式推理任务中失败，因为``幻觉雪球效应``现象，即模型在自由文本反思中递归地为早期错误辩护。尽管结构反馈可以缓解这一问题，现有方法通常依赖外部训练的批评者或符号工具，减少代理自主性。本研究探讨了是否通过基于大纲的约束解码强制结构反思能否在不额外训练的情况下中断错误传播。评估一个80亿参数模型（Qwen3-8B）后发现，单纯施加结构约束并未提升自我校正性能，反而触发了新的失败模式，即``结构雪球效应``。发现满足严格格式规则所需的认知负荷会将模型推入格式陷阱。这一观察解释了为何代理能实现接近完美的表面语法对齐，却无法检测或解决深层语义错误。这些发现揭示了约束解码固有的``对齐税``，突显了结构粒度与内部模型容量在自主工作流中的张力。代码和原始日志可在GitHub仓库中获取：https://github.com/hongxuzhou/agentic_llm_structured_self_critique。
英文摘要：
Intrinsic self-correction in Large Language Models (LLMs) frequently fails in open-ended reasoning tasks due to ``hallucination snowballing,'' a phenomenon in which models recursively justify early errors during free-text reflection. While structured feedback can mitigate this issue, existing approaches often rely on externally trained critics or symbolic tools, reducing agent autonomy. This study investigates whether enforcing structured reflection purely through Outlines-based constrained decoding can disrupt error propagation without additional training. Evaluating an 8-billion-parameter model (Qwen3-8B), we show that simply imposing structural constraints does not improve self-correction performance. Instead, it triggers a new failure mode termed ``structure snowballing.'' We find that the cognitive load required to satisfy strict formatting rules pushes the model into formatting traps. This observation helps explain why the agent achieves near-perfect superficial syntactic alignment yet fails to detect or resolve deeper semantic errors. These findings expose an ``alignment tax'' inherent to constrained decoding, highlighting a tension between structural granularity and internal model capacity in autonomous workflows. Code and raw logs are available in the GitHub repository: https://github.com/hongxuzhou/agentic_llm_structured_self_critique.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
