Guarded Repair for Harm-Aware Post-hoc Replacement of LLM Mathematical Reasoning

• URL: http://arxiv.org/abs/2605.24613v1
• Date: Sat, 23 May 2026 14:51:13 GMT
• Title: Guarded Repair for Harm-Aware Post-hoc Replacement of LLM Mathematical Reasoning
• Abstract summary: GuardedRepair is a best-of-N repair framework that diagnoses cached reasoning traces, selectively triggers repair, and accepts answer-changing candidates.<n>We show that GuardedRepair substantially improves the fixed/broken tradeoff, while also revealing that replacement risk is reduced rather than eliminated.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Post-hoc repair of LLM mathematical reasoning introduces an asymmetric risk: fixing an incorrect reasoning trace is useful, but replacing a trace that was already correct can be harmful. We study this problem under a selective replacement setting, where a system must decide whether a repaired candidate is safer than preserving the original cached trace. We present GuardedRepair, a guarded best-of-N repair framework that diagnoses cached reasoning traces, selectively triggers repair, and accepts answer-changing candidates only when deterministic verification guards support replacement. The framework combines lightweight symbolic checks, surface semantic-risk diagnostics, bounded candidate generation, and conservative acceptance policies. On the full GSM8K test set, where the initial reasoner already achieves 95.60% accuracy, GuardedRepair improves final accuracy to 96.89%, fixing 17 of 58 remaining errors without measured broken-correct cases in the main run. On a weak-reasoner ASDiv setting, accuracy improves from 78.40% to 87.60%. Direct regeneration baselines show that this gain is not explained by stronger-model re-solving alone: re-solving all GSM8K examples lowers accuracy to 93.03% and breaks 47 initially correct answers. Additional analyses show that guarded repair substantially improves the fixed/broken tradeoff, while also revealing that replacement risk is reduced rather than eliminated. These results support viewing post-hoc repair as harm-aware selective replacement rather than unconstrained re-solving.

Related papers

• ReNFT: Repairing Mode Collapse in Reward Post-Training via Internal Probability-Mass Recalibration [84.10540890311843]
  Reward post-training inevitably concentrates probability mass on a few reward-favored modes, a mode collapse that erases within-prompt diversity.<n>We propose ReNFT, which repairs a high-reward, low-diversity adapter through internal probability-mass recalibration.<n>On PickScore and GenEval, ReNFT retains 98.9% and 99.0% of NFT's reward while improving DreamSim-Div by 58.8% and 55.0%, respectively.
  arXiv  Detail & Related papers  (2026-08-30T14:13:50Z)
• AERA: Adaptive Evidence Residual Allocation for Efficient Test-Time Reasoning [22.964476507912327]
  Test-time scaling improves language-model reasoning by generating additional candidate solutions, but allocating the same inference budget to every problem is computationally wasteful.<n>We show that checkpoint-level correctness evolves non-monotonically, and observable evidence may strengthen before an answer collapses or weaken before it recovers.<n>We introduce Adaptive Evidence Residual Allocation (AERA), a sequential controller that learns whether additional computation is likely to recover a better answer from checkpoint-observable evidence.
  arXiv  Detail & Related papers  (2026-08-28T06:21:07Z)
• Beyond Fail-to-Pass: Iterative Hardening of Co-Generated Bug Reproduction Tests and Fixes [55.114939648705395]
  Large language models (LLMs) have made automated program repair increasingly practical for real-world bugs.<n>Bug reproduction tests (BRTs) help close this gap by turning a bug report into an executable, bug-specific signal.<n>We propose CoHarden, a co-generation framework that uses the Lax signal as an in-loop convergence criterion.
  arXiv  Detail & Related papers  (2026-07-22T07:30:07Z)
• Verify, Repair, Repeat, or Stop? Robust Stopping for Noisy Verify-Repair Loops in LLM Agents [8.346035588810516]
  We propose VRR-Stop, a robust stopping framework for noisy verify-repair-repeat loops.<n>VRR-Stop improves final true validity by 60.6 percentage points over fixed five-round repair at an average cost of 0.72 repair rounds.
  arXiv  Detail & Related papers  (2026-07-20T07:52:36Z)
• Conformal Certification of Reasoning Trace Prefixes [30.432255322700126]
  We introduce CROP (Conformal Reasoning Output Prefixes), a verifier-agnostic calibration procedure for clean- annotated certification.<n>We demonstrate that CROP over- and under-withholding, improving downstream repair accuracy by preserving valid intermediate reasoning.
  arXiv  Detail & Related papers  (2026-05-28T15:31:42Z)
• Relative Repairability: A Calibration-Based Diagnostic for High-Sparsity Post-Pruning Allocation [12.288779306467768]
  Relative Repairability (RR) is a calibration based diagnostic that compares the raw activation distortion caused by layerwise pruning with the residual distortion left after channelwise variance matching repair.<n>RR is most useful near an architecture dependent recoverability transition, where standard structural or magnitude based allocation priors begin to lose reliability but post repair recovery has not yet fully collapsed.<n>These results suggest that high sparsity pruning should allocate not only retained weights, but also repairable damage.
  arXiv  Detail & Related papers  (2026-05-25T07:12:38Z)
• MEMOREPAIR: Barrier-First Cascade Repair in Agentic Memory [6.860488391361694]
  We present a barrier-first cascade-repair contract for agentic memory.<n>A repair event induces a controlled transition from invalidated descendant state to validated successor state.<n>We show that the induced publication problem reduces to maximum-weight predecessor closure and can be solved exactly by a single s-t min-cut.
  arXiv  Detail & Related papers  (2026-05-08T04:57:29Z)
• ANNEAL: Adapting LLM Agents via Governed Symbolic Patch Learning [12.44509691077682]
  We introduce ANNEAL, a neuro-symbolic agent that converts recurring failures into governed symbolic edits of a process knowledge graph.<n>Its core mechanism, Failure-Driven Knowledge Acquisition (FDKA), localizes the responsible operator, synthesizes a typed patch through constrained LLM generation, and validates the proposal.<n>Across four domains and 27 multi-seed runs, ANNEAL is the only evaluated system that commits persistent structural repairs.
  arXiv  Detail & Related papers  (2026-05-04T05:24:03Z)
• Are Benchmark Tests Strong Enough? Mutation-Guided Diagnosis and Augmentation of Regression Suites [49.16055123488827]
  Insufficiently strong test suites can admit plausible yet semantically incorrect patches, inflating reported success rates.<n>We introduce STING, a framework for targeted test augmentation that uses semantically altered program variants as diagnostic stressors.<n>STING produces 1,014 validated tests spanning 211 instances and increases patch-region line and branch coverage by 10.8% and 9.5%.
  arXiv  Detail & Related papers  (2026-04-02T01:13:40Z)
• Safety Recovery in Reasoning Models Is Only a Few Early Steering Steps Away [97.11976870616273]
  We propose a lightweight inference-time defense that treats safety recovery as a satising constraint rather than an objective.<n>In our evaluations across six open-source MLRMs and four jailbreak benchmarks, SafeThink reduces attack success rates by 30-60%.
  arXiv  Detail & Related papers  (2026-02-11T18:09:17Z)
• Specification Vibing for Automated Program Repair [8.68148153927532]
  VibeRepair is a specification-centric APR technique that treats repair as behavior-specification repair rather than ad-hoc code editing.<n>On Defects4J v1.2, VibeRepair correctly repairs 174 bugs, exceeding the strongest state-of-the-art baseline by 28 bugs.<n>On Defects4J v2.0, it repairs 178 bugs, outperforming prior approaches by 33 bugs, representing a 23% improvement.
  arXiv  Detail & Related papers  (2026-02-09T04:44:58Z)
• Trusted Uncertainty in Large Language Models: A Unified Framework for Confidence Calibration and Risk-Controlled Refusal [31.458406135473805]
  We present UniCR, a unified framework that turns heterogeneous uncertainty evidence into a calibrated probability of correctness.<n>UniCR learns a lightweight calibration head with temperature scaling and proper scoring.<n>Experiments on short-form QA, code generation with execution tests, and retrieval-augmented long-form QA show consistent improvements in calibration metrics.
  arXiv  Detail & Related papers  (2025-09-01T13:14:58Z)
• Specification-Guided Repair of Arithmetic Errors in Dafny Programs using LLMs [79.74676890436174]
  We present an APR tool for Dafny that uses formal specifications as oracles for fault localization and repair.<n>We localize faults through a series of steps, which include using Hoare logic to determine the state of each statement within the program.<n>Our tool achieves 89.6% fault localization coverage and GPT-4o mini yields the highest repair success rate of 74.18%.
  arXiv  Detail & Related papers  (2025-07-04T15:36:12Z)
• COIN: Uncertainty-Guarding Selective Question Answering for Foundation Models with Provable Risk Guarantees [51.5976496056012]
  COIN is an uncertainty-guarding selection framework that calibrates statistically valid thresholds to filter a single generated answer per question.<n>COIN estimates the empirical error rate on a calibration set and applies confidence interval methods to establish a high-probability upper bound on the true error rate.<n>We demonstrate COIN's robustness in risk control, strong test-time power in retaining admissible answers, and predictive efficiency under limited calibration data.
  arXiv  Detail & Related papers  (2025-06-25T07:04:49Z)
• Latent Veracity Inference for Identifying Errors in Stepwise Reasoning [78.29317733206643]
  We introduce Veracity Search (VS), a discrete search algorithm over veracity assignments.<n>It performs otherwise intractable inference in the posterior distribution over latent veracity values.<n>It generalizes VS, enabling accurate zero-shot veracity inference in novel contexts.
  arXiv  Detail & Related papers  (2025-05-17T04:16:36Z)
• Beyond calibration: estimating the grouping loss of modern neural networks [68.8204255655161]
  Proper scoring rule theory shows that given the calibration loss, the missing piece to characterize individual errors is the grouping loss. We show that modern neural network architectures in vision and NLP exhibit grouping loss, notably in distribution shifts settings.
  arXiv  Detail & Related papers  (2022-10-28T07:04:20Z)

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.