the-producing-agent-should-not-be-sole-judge-of-its-outputmechanismmechanism reasoningpending review
The agent that produced an artifact is a biased judge of it — it holds the context and the incentives that skew its assessment — so verification belongs with a deterministic sensor or a separate verifier that reports failures back rather than rewriting the output, and in a multi-agent system the verifier is the one component no agent may override.
Ingested from a paper but not yet reviewed by a human. It is deliberately inert: it does not move any technique’s standing, does not count toward the backtest, and is excluded anywhere a claim would carry weight. Read the source before relying on it.
Evidence for: No one should be judge in their own cause (holds), Two heads are better than one (narrows)
Capability: Fixing its own mistakes · Agentic
Sources
- Section XVII.C, stated as a principle with the reasoning but no measurement.
- The measured version of the same point on a single model: self-review without an external signal does not improve reasoning. Producer bias is the mechanism this catalog already holds for why.
- Frames sensors as the checks that catch what the agent's own judgment will not, and argues for self-correction loops that run before human eyes precisely because the agent cannot be trusted to grade itself.
Disagreeing is the most useful thing you can do here. Both sides of every contested claim in this catalog were assembled by the same person, which is its weakest point.
Related claims
- In an agent's verify-and-fix loop, deterministic checks — tests, linters, schema validators — should be exhausted before any model-based judge is added, because they are free, fast and give the same verdict every time, while an LLM-as-judge costs tokens per run, returns non-deterministic verdicts, and so belongs only on properties no rule can express, and even then as an advisory signal until measured against human review.Fixing its own mistakes · unreviewed
- In multi-agent LLM systems where one agent starts with an erroneous shared belief, broadcasting every agent's full context at every step raises the rate of false assertions above doing no synchronization at all, because the error propagates to agents that were previously correct — and the harm appears only in tasks where one wrong fact cascades across semantically linked dimensions (destination to airport to weather), not where agent contexts are largely orthogonal.Stating false facts confidently · unreviewed
- Reflect-and-retry raises task success when the feedback in the loop is a genuine external signal — a failing test, a compiler error, an environment outcome — but not when the "feedback" is the model's own unaided critique. The grounding, not the reflection step, is what does the work.Fixing its own mistakes · contested
- Without an external, ground-truth signal — a failing test, a compiler error, a verifier's output — a model's own critique of its reasoning is not a reliable improvement signal, and asking it to review and revise a correct answer often turns it into a wrong one.Fixing its own mistakes
- For persistent-memory agents, any memory-write rule that decides using only recency and provenance is stuck on a single tradeoff — accepting more genuine preference updates means admitting more poisoned ones — because an adversary who can launder a claim through the user's own channel matches the statistics a genuine revision produces; conditioning the write decision on the inferred authenticity of the claim given the interaction history moves both axes at once.Remembering across sessions · unreviewed
Notes
Supports external-feedback-repair-works-only-with-real-grounding from the multi-agent side. Filed separately because "who verifies" is a design decision in its own right, distinct from "does reflection help".