Physically Verifiable Evidence and LLM-Based Reporting for Bearing Fault Diagnosis
Summary
This paper introduces the Diagnostic Evidence Network (DENet), a multi-task framework that extends AI-based bearing fault diagnosis to produce physically verifiable evidence, such as predicted characteristic frequencies and temporal localization of impulses, while using a QLoRA-adapted language model to generate constrained diagnostic reports that reduce hallucinated content.
View Cached Full Text
Cached at: 07/28/26, 06:22 AM
# Physically Verifiable Evidence and LLM-Based Reporting for Bearing Fault Diagnosis Source: [https://arxiv.org/abs/2607.22797](https://arxiv.org/abs/2607.22797) [View PDF](https://arxiv.org/pdf/2607.22797) > Abstract:Trustworthy deployment of AI\-based diagnosis in safety\-critical mechanical systems hinges on validation: whether a prediction can be checked against physical reality before it is acted upon\. Current intelligent fault diagnosers fail this standard in two ways\. Their standard output, a class label with a softmax confidence score, is an internal statistic of the classifier, offering nothing checkable against independent physical knowledge; and the growing use of generative language models in maintenance reporting adds a second risk: hallucinated content entering reports on which decisions rest\. Taking bearing fault diagnosis as the testbed, this work addresses both problems from the output side\. The proposed Diagnostic Evidence Network \(DENet\) is an encoder\-agnostic multi\-task framework extending the output to a structured evidence record: the classification, a predicted characteristic frequency comparable against the theoretical value determined by bearing geometry and shaft speed, and a temporal localization of transient impulses inspectable on the raw waveform\. Across four encoders and three public datasets, this evidence incurs no statistically significant accuracy cost, with a frequency error of about 6 Hz on 1,024\-point segments where spectral estimation is structurally inapplicable\. Centrally, the deviation between predicted and theoretical frequency constitutes a label\-free, inference\-time validation signal: it detects misclassifications with AUROC values of 0\.970 and 0\.871, and remains discriminative in the high\-confidence regime where confidence\-derived detectors are blind\. Finally, a QLoRA\-adapted language model is constrained to translate, but never generate, diagnostic content, reducing unsupported\-claim rates from 10\-12% to 2% and eliminating fabricated quantities\. ## Submission history From: Yuntong Chen \[[view email](https://arxiv.org/show-email/b705868f/2607.22797)\] **\[v1\]**Fri, 24 Jul 2026 14:09:16 UTC \(6,109 KB\)
Similar Articles
Neuro-Symbolic Verification of LLM Outputs for Data-Sensitive Domains (extended preprint)
This paper presents a neuro-symbolic verification architecture for LLM outputs in high-stakes domains, combining formal symbolic methods with neural semantic analysis. Evaluated on a medical device damage assessment system, it achieves over 83% hallucination detection for structured entities and 30% reduction in report creation time.
Evidence-Ledger Adjudication for Claim-Evidence Traceability
This paper introduces evidence-ledger adjudication, a workflow for claim-evidence traceability in AI-assisted writing, evaluated on a blind benchmark from AVeriTeC, CLIMATE-FEVER, and SciFact, showing agent-based methods outperform baselines.
Evidence-Grounded Ensemble Diagnosis of 802.11 Packet Captures: A Multi-Stage Pipeline with Deterministic Reliability Scoring
This paper introduces PROBE, a multi-stage pipeline for diagnosing 802.11 packet captures that combines deterministic normalization, multi-run ensemble, and a verdict-aware evidence framework to produce reliable and calibrated diagnoses, outperforming single-pass LLM analysis and naive ensemble voting.
Dep-LLM: Training-Free Depression Diagnosis via Evidence-Guided Structured Multi-factor with Reliable LLM Reasoning
Dep-LLM is a training-free framework that uses frozen large language models to diagnose depression from clinical interviews by decomposing dialogue into five clinically aligned themes with evidence-grounded reasoning and confidence modulation, outperforming zero-shot and some supervised methods on DAIC-WOZ and E-DAIC datasets.
Reinforcement Learning for Evidence-Seeking Diagnostic Reasoning with Large Language Models
This paper proposes a reinforcement learning framework for evidence-seeking diagnostic reasoning using LLMs. The RL-trained 7B model outperforms larger models in multilingual clinical consultation tasks, showing that specialized RL can distill high-level clinical reasoning.