Screen Before You Interpret: A Portable Validity Protocol for Benchmark-Based LLM Confidence Signals

• URL: http://arxiv.org/abs/2604.17714v1
• Date: Mon, 20 Apr 2026 01:50:38 GMT
• Title: Screen Before You Interpret: A Portable Validity Protocol for Benchmark-Based LLM Confidence Signals
• Abstract summary: LLM confidence signals are used for abstention, routing, and safety-critical decisions.<n>No standard practice exists for checking whether a confidence signal carries item-level information before building on it.<n>We transfer the validity screening principle from clinical personality assessment (PAI, MMPI-3) as a portable protocol for benchmark-based LLM confidence data.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: LLM confidence signals are used for abstention, routing, and safety-critical decisions. No standard practice exists for checking whether a confidence signal carries item-level information before building on it. We transfer the validity screening principle from clinical personality assessment (PAI, MMPI-3) as a portable protocol for benchmark-based LLM confidence data. The protocol specifies three core indices (L, Fp, RBS), a structural indicator (TRIN), and an item-sensitivity statistic, computed from a single 2x2 contingency table. A three-tier classification system (Invalid, Indeterminate, Valid) draws on four clinical traditions. Validated on 20 frontier LLMs across 524 items, four models are classified Invalid, two Indeterminate. Valid-profile models show mean r = .18 (15/16 significant). Invalid-profile models show mean r = -.20 (d = 2.48). Cross-benchmark validation on 18 models using MMLU with verbalized confidence and on external data from Yang et al. (2024) confirms the screen transfers across benchmarks and probe formats. All data and code: https://github.com/synthiumjp/validity-scaling-llm

Related papers

• NovGauge: A Fine-Grained Benchmark for Diagnosing LLMs' Capability in Paper Novelty Assessment [54.4265627247104]
  Large language models (LLMs) are increasingly used in peer review at major AI conferences.<n>Existing benchmarks assess novelty as a single holistic score.<n>We present NovGauge, a human-anchored benchmark for fine-grained novelty assessment diagnosis.
  arXiv  Detail & Related papers  (2026-09-10T08:34:56Z)
• PROTON: Prototype-Based Test-Time Online OOD Detection for Medical VLMs [26.105088397249034]
  Out-of-distribution (OOD) inputs at deployment remain an open problem for medical vision-language models (VLMs)<n>We propose PROTON (PROtotype-based Test-time ONline OOD detection), a lightweight post-hoc module that fuses prototype distance with OOD scoring via stream-level variance statistics.<n>On the benchmark FLAIR + FIVES, PROTON improves MCM by +23.9 AUROC on covariate shift, +8.8 on semantic shift, and +8.1 on far-OOD, making it the only zero-shot method to improve all three without hierarchical prompts or labeled data.
  arXiv  Detail & Related papers  (2026-06-18T20:09:13Z)
• VERDI: Single-Call Confidence Estimation for Verification-Based LLM Judges via Decomposed Inference [0.0]
  VERDI (VERification-Decomposed Inference Inference) is a method that extracts confidence from a structured judge.<n> VERDI decomposes each evaluation into sub-checks and derives three structural signals.<n>On three public benchmarks, VERDI achieves AUROC 0.72-0.91 on GPT-4.1-mini and 0.66-0.80 on GPT-5.4-mini.
  arXiv  Detail & Related papers  (2026-05-11T23:39:19Z)
• Confidence-Aware Alignment Makes Reasoning LLMs More Reliable [65.44962502963378]
  We introduce CASPO, a framework that aligns token-level confidence with step-wise logical correctness without training a separate reward model.<n>During inference, we propose Confidence-aware Thought (CaT), which dynamically prunes uncertain reasoning branches with negligible O(V) latency.<n>Experiments across ten benchmarks and multiple model families show that CASPO consistently improves reasoning reliability and inference efficiency.
  arXiv  Detail & Related papers  (2026-05-08T07:08:25Z)
• Trust but Verify: Introducing DAVinCI -- A Framework for Dual Attribution and Verification in Claim Inference for Language Models [53.279391576560755]
  Large Language Models (LLMs) have demonstrated remarkable fluency and versatility across a wide range of NLP tasks, yet they remain prone to factual inaccuracies and limitation.<n>This poses significant risks in high-stakes domains such as healthcare, law, and scientific communication, where trust and verifiability are paramount.<n>We introduce DAVinCI - a Dual Attribution and Verification framework designed to enhance the factual reliability and interpretability of LLM outputs.
  arXiv  Detail & Related papers  (2026-04-23T01:37:50Z)
• Before You Interpret the Profile: Validity Scaling for LLM Metacognitive Self-Report [0.0]
  We apply the validity scaling framework from the PAI and MMPI-3 to metacognitive probe data.<n>Six validity indices are operationalised: L (maintaining confidence on errors), K (betting on errors), F (withdrawing consensus-endorsed items), Fp (withdrawing correct answers), RBS (inverted monitoring) and TRIN.
  arXiv  Detail & Related papers  (2026-04-20T01:42:54Z)
• NANOZK: Layerwise Zero-Knowledge Proofs for Verifiable Large Language Model Inference [0.0]
  We present METHOD, a zero-knowledge proof system that makes LLM inference verifiable.<n>Our approach exploits the fact that transformer inference naturally decomposes into independent layer computations.<n>Compared to EZKL, METHOD achieves 70x smaller proofs and 5.7x faster proving time at d=128, while maintaining formal soundness guarantees.
  arXiv  Detail & Related papers  (2026-03-17T04:14:45Z)
• Automated Self-Testing as a Quality Gate: Evidence-Driven Release Management for LLM Applications [51.56484100374058]
  We present an automated self-testing framework that introduces quality gates with evidence-based release decisions.<n>We evaluate the framework through a longitudinal case study of an internally deployed multi-agent conversational AI system.
  arXiv  Detail & Related papers  (2026-03-13T20:44:15Z)
• Replayable Financial Agents: A Determinism-Faithfulness Assurance Harness for Tool-Using LLM Agents [0.7699235580548228]
  LLM agents struggle with regulatory audit replay: when asked to reproduce a transaction flagged decision with identical inputs, most deployments fail to return consistent results.<n>This paper introduces the DeterminismFaithfulness Assurance Harness (DFAH), a framework for measuring trajectory determinism and evidence-conditioned faithfulness in tool-using agents deployed in financial services.
  arXiv  Detail & Related papers  (2026-01-17T19:47:55Z)
• Trustworthiness Calibration Framework for Phishing Email Detection Using Large Language Models [0.0]
  Large language models (LLMs) such as GPT-4 and LLaMA-3-8B achieve strong accuracy in text classification.<n>This study introduces the Trustworthiness Framework (TCF), a reproducible methodology for evaluating phishing detectors.
  arXiv  Detail & Related papers  (2025-11-06T18:14:44Z)
• CLUE: Non-parametric Verification from Experience via Hidden-State Clustering [64.50919789875233]
  We show that correctness of a solution is encoded as a geometrically separable signature within the trajectory of hidden activations.<n>ClUE consistently outperforms LLM-as-a-judge baselines and matches or exceeds modern confidence-based methods in reranking candidates.
  arXiv  Detail & Related papers  (2025-10-02T02:14:33Z)
• MedOmni-45°: A Safety-Performance Benchmark for Reasoning-Oriented LLMs in Medicine [69.08855631283829]
  We introduce Med Omni-45 Degrees, a benchmark designed to quantify safety-performance trade-offs under manipulative hint conditions.<n>It contains 1,804 reasoning-focused medical questions across six specialties and three task types, including 500 from MedMCQA.<n>Results show a consistent safety-performance trade-off, with no model surpassing the diagonal.
  arXiv  Detail & Related papers  (2025-08-22T08:38:16Z)
• Self-Evaluation Improves Selective Generation in Large Language Models [54.003992911447696]
  We reformulate open-ended generation tasks into token-level prediction tasks. We instruct an LLM to self-evaluate its answers. We benchmark a range of scoring methods based on self-evaluation.
  arXiv  Detail & Related papers  (2023-12-14T19:09:22Z)
• Factcheck-Bench: Fine-Grained Evaluation Benchmark for Automatic Fact-checkers [121.53749383203792]
  We present a holistic end-to-end solution for annotating the factuality of large language models (LLMs)-generated responses. We construct an open-domain document-level factuality benchmark in three-level granularity: claim, sentence and document. Preliminary experiments show that FacTool, FactScore and Perplexity are struggling to identify false claims.
  arXiv  Detail & Related papers  (2023-11-15T14:41:57Z)
• Assessing the Reliability of Large Language Model Knowledge [78.38870272050106]
  Large language models (LLMs) have been treated as knowledge bases due to their strong performance in knowledge probing tasks. How do we evaluate the capabilities of LLMs to consistently produce factually correct answers? We propose MOdel kNowledge relIabiliTy scORe (MONITOR), a novel metric designed to directly measure LLMs' factual reliability.
  arXiv  Detail & Related papers  (2023-10-15T12:40:30Z)

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.