Skip to content
← Back to explorer

MedDialBench: Benchmarking LLM Diagnostic Robustness under Parametric Adversarial Patient Behaviors

Xiaotian Luo, Xun Jiang, Jiangcheng Wu · Apr 8, 2026 · Citations: 0

Automatic Metrics
Open arXiv Find Implementation RSS feed Shortlist (0)

Data freshness

Extraction: Fresh

Check recency before relying on this page for active eval decisions. Use stale pages as context and verify against current hub results.

Metadata refreshed

Apr 8, 2026, 9:09 AM

Fresh

Extraction refreshed

Apr 10, 2026, 7:11 AM

Fresh

Extraction source

Persisted extraction

Confidence 0.45

Abstract

Interactive medical dialogue benchmarks have shown that LLM diagnostic accuracy degrades significantly when interacting with non-cooperative patients, yet existing approaches either apply adversarial behaviors without graded severity or case-specific grounding, or reduce patient non-cooperation to a single ungraded axis, and none analyze cross-dimension interactions. We introduce MedDialBench, a benchmark enabling controlled, dose-response characterization of how individual patient behavior dimensions affect LLM diagnostic robustness. It decomposes patient behavior into five dimensions -- Logic Consistency, Health Cognition, Expression Style, Disclosure, and Attitude -- each with graded severity levels and case-specific behavioral scripts. This controlled factorial design enables graded sensitivity analysis, dose-response profiling, and cross-dimension interaction detection. Evaluating five frontier LLMs across 7,225 dialogues (85 cases x 17 configurations x 5 models), we find a fundamental asymmetry: information pollution (fabricating symptoms) produces 1.7-3.4x larger accuracy drops than information deficit (withholding information), and fabricating is the only configuration achieving statistical significance across all five models (McNemar p < 0.05). Among six dimension combinations, fabricating is the sole driver of super-additive interaction: all three fabricating-involving pairs produce O/E ratios of 0.70-0.81 (35-44% of eligible cases fail under the combination despite succeeding under each dimension alone), while all non-fabricating pairs show purely additive effects (O/E ~ 1.0). Inquiry strategy moderates deficit but not pollution: exhaustive questioning recovers withheld information, but cannot compensate for fabricated inputs. Models exhibit distinct vulnerability profiles, with worst-case drops ranging from 38.8 to 54.1 percentage points.

Low-signal caution for protocol decisions

Use this page for context, then validate protocol choices against stronger HFEPX references before implementation decisions.

  • Extraction flags indicate low-signal or possible false-positive protocol mapping.
  • Extraction confidence is 0.45 (below strong-reference threshold).
Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub

HFEPX Relevance Assessment

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Best use

Background context only

Use if you need

A benchmark-and-metrics comparison anchor.

Main weakness

Extraction flags indicate low-signal or possible false-positive protocol mapping.

Trust level

Low

Eval-Fit Score

5/100 • Low

Treat as adjacent context, not a core eval-method reference.

Human Feedback Signal

Not explicit in abstract metadata

Evaluation Signal

Detected

HFEPX Fit

Adjacent candidate

Extraction confidence: Low

If you are doing eval pipeline work, start here:

Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub Tool-Use Eval Hub

Field Provenance & Confidence

Each key protocol field shows extraction state, confidence band, and data source so you can decide whether to trust it directly or validate from full text.

Human Feedback Types

missing

None explicit

Confidence: Low Source: Persisted extraction missing

No explicit feedback protocol extracted.

Evidence snippet: Interactive medical dialogue benchmarks have shown that LLM diagnostic accuracy degrades significantly when interacting with non-cooperative patients, yet existing approaches either apply adversarial behaviors without graded severity or case-specific grounding, or reduce patient non-cooperation to a single ungraded axis, and none analyze cross-dimension interactions.

Evaluation Modes

partial

Automatic Metrics

Confidence: Low Source: Persisted extraction evidenced

Includes extracted eval setup.

Evidence snippet: Interactive medical dialogue benchmarks have shown that LLM diagnostic accuracy degrades significantly when interacting with non-cooperative patients, yet existing approaches either apply adversarial behaviors without graded severity or case-specific grounding, or reduce patient non-cooperation to a single ungraded axis, and none analyze cross-dimension interactions.

Quality Controls

missing

Not reported

Confidence: Low Source: Persisted extraction missing

No explicit QC controls found.

Evidence snippet: Interactive medical dialogue benchmarks have shown that LLM diagnostic accuracy degrades significantly when interacting with non-cooperative patients, yet existing approaches either apply adversarial behaviors without graded severity or case-specific grounding, or reduce patient non-cooperation to a single ungraded axis, and none analyze cross-dimension interactions.

Benchmarks / Datasets

partial

Meddialbench

Confidence: Low Source: Persisted extraction evidenced

Useful for quick benchmark comparison.

Evidence snippet: We introduce MedDialBench, a benchmark enabling controlled, dose-response characterization of how individual patient behavior dimensions affect LLM diagnostic robustness.

Reported Metrics

partial

Accuracy

Confidence: Low Source: Persisted extraction evidenced

Useful for evaluation criteria comparison.

Evidence snippet: Interactive medical dialogue benchmarks have shown that LLM diagnostic accuracy degrades significantly when interacting with non-cooperative patients, yet existing approaches either apply adversarial behaviors without graded severity or case-specific grounding, or reduce patient non-cooperation to a single ungraded axis, and none analyze cross-dimension interactions.

Rater Population

missing

Unknown

Confidence: Low Source: Persisted extraction missing

Rater source not explicitly reported.

Evidence snippet: Interactive medical dialogue benchmarks have shown that LLM diagnostic accuracy degrades significantly when interacting with non-cooperative patients, yet existing approaches either apply adversarial behaviors without graded severity or case-specific grounding, or reduce patient non-cooperation to a single ungraded axis, and none analyze cross-dimension interactions.

Human Data Lens

  • Uses human feedback: No
  • Feedback types: None
  • Rater population: Unknown
  • Unit of annotation: Unknown
  • Expertise required: Medicine
  • Extraction source: Persisted extraction

Evaluation Lens

  • Evaluation modes: Automatic Metrics
  • Agentic eval: None
  • Quality controls: Not reported
  • Confidence: 0.45
  • Flags: low_signal, possible_false_positive

Protocol And Measurement Signals

Benchmarks / Datasets

Meddialbench

Reported Metrics

accuracy

Research Brief

Deterministic synthesis

Interactive medical dialogue benchmarks have shown that LLM diagnostic accuracy degrades significantly when interacting with non-cooperative patients, yet existing approaches either apply adversarial behaviors without graded severity or… HFEPX signals include Automatic Metrics with confidence 0.45. Updated from current HFEPX corpus.

Generated Apr 10, 2026, 7:11 AM · Grounded in abstract + metadata only

Key Takeaways

  • Interactive medical dialogue benchmarks have shown that LLM diagnostic accuracy degrades significantly when interacting with non-cooperative patients, yet existing approaches…
  • We introduce MedDialBench, a benchmark enabling controlled, dose-response characterization of how individual patient behavior dimensions affect LLM diagnostic robustness.

Researcher Actions

  • Treat this as method context, then pivot to protocol-specific HFEPX hubs.
  • Cross-check benchmark overlap: Meddialbench.
  • Validate metric comparability (accuracy).

Caveats

  • Generated from title, abstract, and extracted metadata only; full-paper implementation details are not parsed.
  • Low-signal flag detected: protocol relevance may be indirect.

Research Summary

Contribution Summary

  • Interactive medical dialogue benchmarks have shown that LLM diagnostic accuracy degrades significantly when interacting with non-cooperative patients, yet existing approaches either apply adversarial behaviors without graded severity or…
  • We introduce MedDialBench, a benchmark enabling controlled, dose-response characterization of how individual patient behavior dimensions affect LLM diagnostic robustness.
  • Evaluating five frontier LLMs across 7,225 dialogues (85 cases x 17 configurations x 5 models), we find a fundamental asymmetry: information pollution (fabricating symptoms) produces 1.7-3.4x larger accuracy drops than information deficit…

Why It Matters For Eval

  • Interactive medical dialogue benchmarks have shown that LLM diagnostic accuracy degrades significantly when interacting with non-cooperative patients, yet existing approaches either apply adversarial behaviors without graded severity or…
  • We introduce MedDialBench, a benchmark enabling controlled, dose-response characterization of how individual patient behavior dimensions affect LLM diagnostic robustness.

Researcher Checklist

  • Gap: Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • Pass: Evaluation mode is explicit

    Detected: Automatic Metrics

  • Gap: Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Pass: Benchmark or dataset anchors are present

    Detected: Meddialbench

  • Pass: Metric reporting is present

    Detected: accuracy

Category-Adjacent Papers (Broader Context)

These papers are nearby in arXiv category and useful for broader context, but not necessarily protocol-matched to this paper.

Need human evaluators for your AI research? Scale annotation with expert AI Trainers.

Post a Job Get a Quote