Deposon：一种可审计、守恒保证、博弈论验证的LLM推理路径散射层
Deposon: An Auditable, Conservation-Guaranteed, Game-Theoretically Tested Scattering Layer over LLM Reasoning Paths
浏览论文内容
中文总结 AI 辅助
提出Deposon散射层，为LLM推理路径提供可审计账本，通过三通道散射保证能量守恒，并经过博弈论验证，其价值在于机器可验证性而非性能提升。
中文摘要 AI 辅助
多步LLM推理缺乏机器可复核的账本：被丢弃的推理路径不留下可审计的记录。我们提出Deposon散射层，它将LLM生成的概念分解图的每个节点绑定到一个双参数Deposon状态；路径经历三通道散射——透射、反射、不可逆耗散——对于任意参数满足T+R+A=1，每条路径的最大能量审计偏差为2.2E-16（机器精度）。我们诚实地报告所有三个证据层级。在合成陷阱基准上，路径过滤增益是闭合的（预注册）：统一达到100%，而诱饵捕获基线为7%/10%。在真实基准上，该层与一个简单的六关键词规则过滤器无法区分（GSM8K 0.87 >= 0.85，McNemar p=0.5；StrategyQA 0.899 = 0.899）；此处未检测到差异，因此我们将主张锐化为“差异价值仅在于机器可验证性”。融合产生第二个负面结果：与语义先验的凸组合从未改进（物理0.484 -> 0.452），且明显的lambda=2增益是反场伪影；任何融合增益必须是非线性的。将反向动力学建模为图上的势博弈，我们证明了可审计标量的单调性和近似梯度性，并量化了经验协调比率（ECR）。三个形式化的动力学等价命题（P1a/P1b/T-P1c）在预注册的终止协议下被证伪，势博弈主张被降级为近似（循环图中位数残差0.669）：在动力学层面只有一致性级别的证据幸存。代码：此HTTP URL。
英文摘要
Multi-step LLM reasoning lacks a machine-recheckable ledger: discarded reasoning paths leave no auditable record. We propose the Deposon scattering layer, which binds each node of an LLM-generated concept-decomposition graph to a two-parameter Deposon state; paths undergo three-channel scattering -- transmission, reflection, irreversible dissipation -- obeying T+R+A=1 for arbitrary parameters, with a maximum per-path energy-audit deviation of 2.2E-16 (machine epsilon). We report all three evidence tiers honestly. On synthetic trap benchmarks the path-filtering gain is closed (pre-registered): unified reaches 100% versus a decoy-capture baseline at 7%/10%. On real benchmarks the layer is indistinguishable from a trivial six-keyword rule filter (GSM8K 0.87 >= 0.85, McNemar p=0.5; StrategyQA 0.899 = 0.899); no difference is detected here, so we sharpen the claim to "the differential value lies solely in machine verifiability." Fusion yields a second negative result: convex combinations with a semantic prior never improve (physics 0.484 -> 0.452), and the apparent lambda=2 gain is an anti-field artifact; any fusion gain must be nonlinear. Modeling the reverse dynamics as a potential game on the graph, we evidence an auditable scalar's monotonicity and near-gradientness and quantify the empirical coordination ratio (ECR). The three formalized dynamical-equivalence propositions (P1a/P1b/T-P1c) are falsified under the pre-registered kill protocol, and the potential-game claim is downgraded to approximate (cyclic-graph median residual 0.669): only consistency-level evidence survives at the dynamical level. Code: github.com/zeroandcat/Deposon.
发表机构
- School of Chemistry and Life Resources, Renmin University of China（中国人民大学化学与生命资源学院）
机构由 AI 辅助整理，请以论文原文为准。
