Failures Reveal What Metrics Miss: An Evidence-Driven Agent for Recursive Refinement of ECG Classifiers

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Jinliang Deng, Yiming Niu, Yibo Pan, Zhiqi Shao, Qin Luo, Yongxin Tong
• Venue: arXiv: cs.AI
• Published: 2026-07-27
• Collected: 2026-07-28
• URL: http://arxiv.org/abs/2607.24419v1
• PDF: https://arxiv.org/pdf/2607.24419v1
• DOI: UNKNOWN

Plain-Language Brief

Researchers present a new framework called RecursiveECG that uses a language model (LLM) to improve ECG classifiers by analyzing the reasons for individual case failures and suggesting targeted revisions based on concrete evidence. The framework is designed to outperform strong baselines in various datasets, achieving an average relative improvement of 10.0%.

What the Document Says

• RecursiveECG is a new framework that uses an LLM for designing ECG classifiers.
• The framework refines ECG classifiers based on concrete failures and objective ECG evidence.
• Criteria-to-Measurement Compilation converts curated ECG criteria into deterministic functions for individual ECGs.
• Evidence-Grounded Failure Review analyzes failed cases by considering raw waveforms, measurements, and model outputs.
• Candidate revisions are executed and re-evaluated under a fixed problem contract.
• RecursiveECG consistently outperforms strong baselines, achieving an average relative improvement of 10.0%.
• The framework is tested on PTB-XL, Georgia, and CPSC2018 datasets.

Key Concepts Mentioned

• RecursiveECG
• LLM
• ECG classifiers
• Criteria-to-Measurement Compilation
• Evidence-Grounded Failure Review
• strong baselines

Claims Made by the Document

• RecursiveECG outperforms strong baselines.
• The average relative improvement of RecursiveECG is 10.0%.

Evidence Used

• Across PTB-XL, Georgia, and CPSC2018, RecursiveECG consistently outperforms strong baselines.
• The study reports an average relative improvement of 10.0%.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Deep models have substantially advanced 12-lead ECG classification, yet their refinement still relies heavily on human experts to inspect failures and iteratively revise classifier designs. Recent LLM-based agents have demonstrated the potential for automated model design, but when guided only by aggregate performance metrics, they lack insight into why individual cases fail and how the classifier should be revised. We present RecursiveECG, an evidence-driven LLM-as-Designer framework in which an LLM serves as an offline model designer that refines ECG classifiers based on concrete failures and objective ECG evidence. To ground failure diagnosis in executable evidence, Criteria-to-Measurement Compilation converts curated ECG criteria into validated deterministic functions that produce reproducible, reference-backed measurements for individual ECGs. Building on these measurements, Evidence-Grounded Failure Review analyzes failed and comparator cases by jointly considering raw waveforms, measurements, and model outputs, enabling the LLM to diagnose classifier limitations and formulate targeted revisions. Candidate revisions are executed and re-evaluated under a fixed problem contract, and only evidence-supported updates are retained. The resulting predictor is frozen after refinement and requires no LLM inference during deployment, while an audit trail links each accepted revision to its supporting evidence. Across PTB-XL, Georgia, and CPSC2018, RecursiveECG consistently outperforms strong baselines, achieving an average relative improvement of 10.0%. Extensive ablation and transfer studies further validate the effectiveness of its evidence-grounded refinement process.