Toward Measuring Structural Drift in LLM Communication Loops

Excerpt

arXiv:2604.13061v3 Announce Type: replace Abstract: Large language models increasingly run in stateful pipelines that assemble each prompt from retrieval, memory, tools, and other agents. Such pipelines drift: information that should shape the next response is dropped, compressed, or misrouted while every component still reports success. Existing diagnostics miss this because they evaluate isolated prompts, responses, or task scores, whereas what decouples is the relation between a prompt and th