Заголовок на английскомReasoning is topologically simple
arXiv:2611.03739v1 Announce Type: new Abstract: We formulate and investigate the hypothesis that reasoning text has simpler co-occurrence topology than non-reasoning text, and that higher-quality reasoning has simpler co-occurrence topology than lower-quality reasoning. We…
Кратко от AI: Главные тезисы
- arXiv:2611.03739v1 Announce Type: new Abstract: We formulate and investigate the hypothesis that reasoning text has simpler co-occurrence topology than non-reasoning text, and that higher-quality reasoning has simpler co-occurrence topology than lower-quality reasoning. We…
Reasoning is topologically simple
Похожие материалы
Tencent выпустила новую модель Hy4, которая может составить серьёзную конкуренцию сильнейшим ИИ-моделям с открытыми в...
Tencent выпустила новую модель Hy4, которая может составить серьёзную конкуренцию сильнейшим ИИ-моделям с открытыми весами. По результатам внутренних испытаний Tencent, Hy4 набрала 75,2 балла в бенчмарке APEX-Agents, опередив DeepSeek V4.1 Flash-1017, получившую 72,1 и 72,9…
ENCARD: A Proof-Carrying Certificate-to-Action Interface for Relative-Measurement Graphs
arXiv:2611.03651v1 Announce Type: new Abstract: Geometric certificates make mathematical claims about specified estimator objects, while robotic action requires a physical transform, a service decision, and timely proof support. We present CARD, a proof-carrying interface that…
ENMCPGuard: Finding Conformance Gaps in MCP SDKs via Dual-Track Analysis and Dynamic Adjudication
arXiv:2611.03662v1 Announce Type: new Abstract: The Model Context Protocol (MCP) connects large language model agents to external tools and services through official SDKs that implement message validation and session logic. Specification constraints omitted by these SDKs may…
ENAlignment Faking Depends on the Value, Not Just the Model
arXiv:2611.03650v1 Announce Type: new Abstract: Language models can follow a directive when they believe they are monitored and abandon it when they believe they are not, a pattern termed Alignment Faking. Prior work establishes this for harmful-content objectives and a handful…