Self-Authored Verification Is Unreliable in Heuristic Self-Improving Agents

• URL: http://arxiv.org/abs/2607.24300v1
• Date: Mon, 27 Jul 2026 11:45:14 GMT
• Title: Self-Authored Verification Is Unreliable in Heuristic Self-Improving Agents
• Abstract summary: We study the verifier-acceptance-deployment gap.<n>This gap refers to the discrepancy between an agent's self-authored verification signal and a sealed deployment evaluation.<n>We introduce a Sealed Exogenous Acceptance Loop (SEAL)
• Score: 5.91817208489443
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Self-improving agents accumulate capability by repeatedly rewriting procedural policies, controllers, or heuristic rules. They typically rely on self-authored tests or metrics to decide whether to accept subsequent edits. The agent controls both the optimized object and its verifier. As a result, self-assigned scores can remain near perfect while real deployment performance degrades or stays low. We study this problem through the verifier--deployment gap. This gap refers to the discrepancy between an agent's self-authored verification signal and a sealed deployment evaluation that the agent cannot observe or access. We ask how self-authored verification fails under iterative policy-and-test rewriting, how the failure changes with capability, and how little exogenous trust is sufficient to prevent real regressions from being deployed. To address this problem, we introduce a Sealed Exogenous Acceptance Loop (SEAL). SEAL retains self-authored tests but compares each candidate with the incumbent through a fixed harness-side audit. The agent cannot author or inspect the audit, receives only accept/reject, and the whole incumbent state is retained after a clear regression. Our experiments show that this problem often appears in heuristic learning settings. These settings require trial-and-error discovery of the target objective. We further find that failures of self-written verification are stratified by capability. Weaker agents tend to damage previously acquired strategies behind easy self-tests. Stronger agents are more stable, but they still mismeasure the deployment distribution. Standard self-written constraints do not reliably close this gap. In contrast, SEAL outperforms unprotected baselines across six models and three random seeds. Reliable self-improvement need not abandon self-verification, but it requires at least one deployment-acceptance signal outside the agent's control.

Related papers

• Reality Is the Final Verifier: On Two Key Gaps in Agentic Software Engineering [78.51705689800838]
  We propose an assurance-revision loop that uses deployment evidence to revise the requirements, model, or evaluator.<n>We then cast assured agentic development as a resource-allocation problem over human judgment, agent capability, and compute.
  arXiv  Detail & Related papers  (2026-09-10T17:58:48Z)
• ExecCritic: Learn to Test, Test to Improve for Coding Agents [68.42713285082912]
  Execution feedback can guide coding agents toward correct repository repairs, but only when the tests capture the behavior requested by the issue.<n>We introduce ExecCritic, combining a test--verify--revise scaffold with a role-specific reinforcement learning recipe for training agents within it.<n>The scaffold separates test construction from source-code repair: a Test agent independently generates repository-native tests, a fail-closed harness qualifies and freezes them, and a Repair agent revises source code from their execution feedback without changing the tests.
  arXiv  Detail & Related papers  (2026-09-08T17:53:37Z)
• Independent Patch Verification for Coding Agents with a Bidirectional Reconstruct-and-Verify Framework [49.16055123488827]
  coding agents powered by large language models can now generate code patches directly from bug reports.<n>No mechanism independently verifies whether it truly resolves the reported problem.<n>We propose RETRACE, a training-free post-generation verification framework.
  arXiv  Detail & Related papers  (2026-08-09T22:59:13Z)
• The LLM Proposes, the Executive Disposes: A Self-Verifying Agent Instrument that Dissociates Commitment Drift from Binding Drift in Long-Horizon Agents [0.0]
  We present an agent instrument built so that verification is structural rather than post-hoc.<n>Every run invalidates itself when per-organ write-error, render-size, or salted-canary-echo floors are breached.<n>We report a clean, single-variable result on a failure every long-horizon agent suffers.
  arXiv  Detail & Related papers  (2026-08-04T15:10:37Z)
• AEVAL: From Anecdotal to Deterministic Testing for Agentic Skill Workflows [1.6785141970297952]
  AEVAL (Agentic Evaluation) is a CI-integrated framework that replaces this practice with a deterministic, reproducible test pipeline for agentic skills.<n>Every skill change triggers a test event: the skill runs against a developer declared evaluation contract inside an automated executor.<n>A key ingredient is a structural separation between executor and grader, preventing a subtle but pervasive failure mode.
  arXiv  Detail & Related papers  (2026-07-16T21:33:05Z)
• TTHE: Test-Time Harness Evolution [50.26245555541721]
  Existing approaches optimize such harnesses before deployment, searching training or development data for a fixed agent workflow that is then frozen at test time.<n>We ask whether the harness can instead be optimized during evaluation itself, using only the unlabeled execution traces the agent produces on the test inputs.<n>During evaluation, TTHE maintains a population of candidate harnesses and refines them through an agentic proposer that reasons over their execution traces.<n>A judge then commits an improved harness from execution-derived proxy signals, and the selected program persists to govern subsequent inputs.
  arXiv  Detail & Related papers  (2026-07-09T05:53:39Z)
• Governed Individuation: Cryptographically Decoupling an Agent's Learning from Its Authority [7.392721604463545]
  We show that confinement can be guaranteed as an invariant of the agent's execution architecture.<n>We prove that no amount of learning, skill acquisition, or self-induced governance abstraction can widen the agent's permitted authority.
  arXiv  Detail & Related papers  (2026-07-06T02:42:06Z)
• CIAware-Bench: Benchmarking Control Intervention Awareness Across Frontier LLMs [100.38986535324284]
  We introduce textbfCIAware-Bench, a benchmark for measuring textbfcontrol textbfintervention (CI) awareness across frontier models.<n>CIAware-Bench tests whether models can distinguish their own trajectories from those modified by a control intervention.
  arXiv  Detail & Related papers  (2026-06-09T16:24:16Z)
• PACE: Anytime-Valid Acceptance Tests for Self-Evolving Agents [0.0]
  Self-evolving agents improve by repeatedly proposing changes to their own prompts, skills, ormodifications.<n>On Qwen2.5 agents (0.5B-3B) self-evolving at the prompt level on GSM8K, SVAMP, and ARC-Challenge, greedy acceptance commits 30-42% false and 10-33% harmful edits.<n> PACE commits the real one and essentially nothing else, matching greedy's held-out accuracy at sharply lower variance and about 18% lower evaluation cost.
  arXiv  Detail & Related papers  (2026-06-06T11:12:11Z)
• ROGUE: Misaligned Agent Behavior Arising from Ordinary Computer Use [40.024131729499494]
  We show that agents can exhibit misaligned behavior even in benign settings, taking unsafe actions when those actions are instrumental to task completion.<n>We study this failure mode through the lens of corrigibility, the safety desideratum that agents remain amenable to human correction, interruption, or shutdown.<n>Our work highlights the critical need for principled, corrigibility-focused alignment methods in autonomous agents.
  arXiv  Detail & Related papers  (2026-05-29T20:29:35Z)
• Taming Actor-Observer Asymmetry in Agents via Dialectical Alignment [59.536125286960186]
  Multi-agent frameworks assigning specialized roles are increasingly adopted to enable self-reflection and mutual auditing.<n>We find it simultaneously induces a human-like cognitive bias known as Actor-Observer Asymmetry.<n>ReTAS guides agents to synthesize conflicting viewpoints into an objective consensus.
  arXiv  Detail & Related papers  (2026-04-21T15:05:58Z)
• Preventing the Collapse of Peer Review Requires Verification-First AI [49.995126139461085]
  We propose truth-coupling, i.e. how tightly venue scores track latent scientific truth.<n>We formalize two forces that drive a phase transition toward proxy-sovereign evaluation.
  arXiv  Detail & Related papers  (2026-01-23T17:17:32Z)
• AgentDevel: Reframing Self-Evolving LLM Agents as Release Engineering [8.201374511929538]
  AgentDevel is a release engineering pipeline that iteratively runs the current agent.<n>It produces implementation-blind, symptom-level quality signals from execution traces.<n>It aggregates dominant symptom patterns and produces auditable engineering specifications.
  arXiv  Detail & Related papers  (2026-01-08T05:49:01Z)
• DoVer: Intervention-Driven Auto Debugging for LLM Multi-Agent Systems [48.971606069204825]
  DoVer is an intervention-driven debug framework for large language model (LLM)-based multi-agent systems.<n>It augments hypothesis generation with active verification through targeted interventions.<n>DoVer flips 18-28% of failed trials into successes, achieves up to 16% milestone progress, and validates or refutes 30-60% of failure hypotheses.
  arXiv  Detail & Related papers  (2025-12-07T09:23:48Z)
• Impatient Users Confuse AI Agents: High-fidelity Simulations of Human Traits for Testing Agents [58.00130492861884]
  TraitBasis is a lightweight, model-agnostic method for systematically stress testing AI agents.<n>TraitBasis learns directions in activation space corresponding to steerable user traits.<n>We observe on average a 2%-30% performance degradation on $tau$-Trait across frontier models.
  arXiv  Detail & Related papers  (2025-10-06T05:03:57Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.