用共识图取代代码，让AI编程可追溯、可审计、更可控。
Scaling Human-AI Coding Collaboration Requires a Governable Consensus Layer
- 构建带类型属性的共识图作为工程主产物，替代传统代码
- 通过同步操作保持代码与共识图一致，降低系统脆弱性
- 引入共识熵等指标评估协作质量，适合需要可解释性的团队
Vibe编程虽能快速生成正确可执行的代码，但缺乏对结构承诺、依赖关系和依据的记录。审查者无法判断假设条件、变更内容或回归原因。这并非生成失败，而是控制失效：当前主流的人工智能辅助开发模式（代码+聊天记录）导致维度坍塌，将复杂系统拓扑压缩为低维文本，使系统在变更下变得不透明且脆弱。本文提出「代理共识」范式，以可操作的世界模型——类型属性图C作为工程核心产物，替代代码。可执行产物从C派生，并通过同步算子Phi（实现）与Psi（重构）保持一致性。证据直接关联到C中的结构化声明，使每项承诺均可审计，未明确定义的部分以可测量的共识熵形式显式表达，而非无声猜测。评估应从代码正确性转向对齐保真度、共识熵与干预距离。我们设计了基准任务族，用于衡量基于共识的工作流是否比基于聊天的基线减少人工干预。
原文摘要 · Abstract (English)
Vibe coding produces correct, executable code at speed, but leaves no record of the structural commitments, dependencies, or evidence behind it. Reviewers cannot determine what invariants were assumed, what changed, or why a regression occurred. This is not a generation failure but a control failure: the dominant artifact of AI-assisted development (code plus chat history) performs dimension collapse, flattening complex system topology into low-dimensional text and making systems opaque and fragile under change. We propose Agentic Consensus: a paradigm in which the consensus layer C, an operable world model represented as a typed property graph, replaces code as the primary artifact of engineering. Executable artifacts are derived from C and kept in correspondence via synchronization operators Phi (realize) and Psi (rehydrate). Evidence links directly to structural claims in C, making every commitment auditable and under-specification explicit as measurable consensus entropy rather than a silent guess. Evaluation must move beyond code correctness toward alignment fidelity, consensus entropy, and intervention distance. We propose benchmark task families designed to measure whether consensus-based workflows reduce human intervention compared to chat-driven baselines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。