Article

Verifier Exploitation in NLI-Guided Iterative Refinement: A Controlled Empirical Analysis

2026-06-30

Abstract excerpt

<title>Abstract</title> <p> Verifier exploitation, the pathology whereby a system satisfies its auditing metric while degrading the property it was meant to measure, has been documented in gradient-based and selection-based systems. We show that neither condition is necessary. Using <italic>AnchorSum</italic> , a modular training-free multi-document summarization pipeline with fully separable auditor, feedback...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 7e319cb7-effe-5550-8860-d94e8a258d8c
DOI: 10.21203/rs.3.rs-10187492/v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Verifier Exploitation in NLI-Guided Iterative Refinement: A Controlled Empirical AnalysisDOI 10.21203/rs.3.rs-10187492/v1Read

Limits of Self-Correction in LLMs: An Information-Theoretic Analysis of Correlated ErrorsSemantic proximity Read

How LLMs Audit Each Other: Five Mechanisms of Auditor Bias in Cross-Model Peer Review Under Identity Disclosure and Cross-Lingual ConditionsSemantic proximity Read

Explainable Automatic Evaluation: Interpretable Decision-Making in Large Language Model AutoratersSemantic proximity Read

A Measurement Note on Pre-Wrap and Reader-Visible Context Accounting for a Capped FLAN-T5 QA PipelineSemantic proximity Read

Signed Layer-Level Evidence in Transformer Predictions, and a Diagnostic for When It AppliesSemantic proximity Read

SAS Multimetric Tribunal: Experimental Validation of a Three-Zone Structural Classification System for Hallucination Detection in AI-Generated TextSemantic proximity Read

RePCS: Diagnosing Data Memorization in LLM-Powered Retrieval-Augmented GenerationSemantic proximity Read

Enhancing Reliability in Large Language Models: Self-Detection of Hallucinations With Spontaneous Self-ChecksSemantic proximity Read

From Illusion to Insight: A Taxonomic Survey of Hallucination Mitigation Techniques in LLMsSemantic proximity Read

Optimizing Human-Controlled Preference Alignment in Large Language Models via Dense Token Masking: A Methodological ApproachSemantic proximity Read

Optimizing Text Summarization with Reinforcement Learning from Propagation Feedback Using Large Language ModelsSemantic proximity Read

Foresight Is Not Enough: Sentence-Level Future Signals, Self-Loop Hard Negatives, and the Calibration Gap in Small Language ModelsSemantic proximity Read