Skip to content
← Back to explorer

When to Call an Apple Red: Humans Follow Introspective Rules, VLMs Don't

Jonathan Nemitz, Carsten Eickhoff, Junyi Jessy Li, Kyle Mahowald, Michal Golovanevsky, William Rudman · Apr 7, 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 7, 2026, 7:59 PM

Recent

Extraction refreshed

Apr 10, 2026, 7:25 AM

Fresh

Extraction source

Persisted extraction

Confidence 0.35

Abstract

Understanding when Vision-Language Models (VLMs) will behave unexpectedly, whether models can reliably predict their own behavior, and if models adhere to their introspective reasoning are central challenges for trustworthy deployment. To study this, we introduce the Graded Color Attribution (GCA) dataset, a controlled benchmark designed to elicit decision rules and evaluate participant faithfulness to these rules. GCA consists of line drawings that vary pixel-level color coverage across three conditions: world-knowledge recolorings, counterfactual recolorings, and shapes with no color priors. Using GCA, both VLMs and human participants establish a threshold: the minimum percentage of pixels of a given color an object must have to receive that color label. We then compare these rules with their subsequent color attribution decisions. Our findings reveal that models systematically violate their own introspective rules. For example, GPT-5-mini violates its stated introspection rules in nearly 60\% of cases on objects with strong color priors. Human participants remain faithful to their stated rules, with any apparent violations being explained by a well-documented tendency to overestimate color coverage. In contrast, we find that VLMs are excellent estimators of color coverage, yet blatantly contradict their own reasoning in their final responses. Across all models and strategies for eliciting introspective rules, world-knowledge priors systematically degrade faithfulness in ways that do not mirror human cognition. Our findings challenge the view that VLM reasoning failures are difficulty-driven and suggest that VLM introspective self-knowledge is miscalibrated, with direct implications for high-stakes deployment.

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 0.35 (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 secondary eval reference to pair with stronger protocol papers.

Main weakness

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

Trust level

Low

Eval-Fit Score

0/100 • 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: Understanding when Vision-Language Models (VLMs) will behave unexpectedly, whether models can reliably predict their own behavior, and if models adhere to their introspective reasoning are central challenges for trustworthy deployment.

Evaluation Modes

partial

Automatic Metrics

Confidence: Low Source: Persisted extraction evidenced

Includes extracted eval setup.

Evidence snippet: Understanding when Vision-Language Models (VLMs) will behave unexpectedly, whether models can reliably predict their own behavior, and if models adhere to their introspective reasoning are central challenges for trustworthy deployment.

Quality Controls

missing

Not reported

Confidence: Low Source: Persisted extraction missing

No explicit QC controls found.

Evidence snippet: Understanding when Vision-Language Models (VLMs) will behave unexpectedly, whether models can reliably predict their own behavior, and if models adhere to their introspective reasoning are central challenges for trustworthy deployment.

Benchmarks / Datasets

missing

Not extracted

Confidence: Low Source: Persisted extraction missing

No benchmark anchors detected.

Evidence snippet: Understanding when Vision-Language Models (VLMs) will behave unexpectedly, whether models can reliably predict their own behavior, and if models adhere to their introspective reasoning are central challenges for trustworthy deployment.

Reported Metrics

partial

Faithfulness

Confidence: Low Source: Persisted extraction evidenced

Useful for evaluation criteria comparison.

Evidence snippet: To study this, we introduce the Graded Color Attribution (GCA) dataset, a controlled benchmark designed to elicit decision rules and evaluate participant faithfulness to these rules.

Rater Population

missing

Unknown

Confidence: Low Source: Persisted extraction missing

Rater source not explicitly reported.

Evidence snippet: Understanding when Vision-Language Models (VLMs) will behave unexpectedly, whether models can reliably predict their own behavior, and if models adhere to their introspective reasoning are central challenges for trustworthy deployment.

Human Data Lens

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

Evaluation Lens

  • Evaluation modes: Automatic Metrics
  • Agentic eval: None
  • Quality controls: Not reported
  • Confidence: 0.35
  • Flags: low_signal, possible_false_positive

Protocol And Measurement Signals

Benchmarks / Datasets

No benchmark or dataset names were extracted from the available abstract.

Reported Metrics

faithfulness

Research Brief

Deterministic synthesis

To study this, we introduce the Graded Color Attribution (GCA) dataset, a controlled benchmark designed to elicit decision rules and evaluate participant faithfulness to these rules. HFEPX signals include Automatic Metrics with confidence 0.35. Updated from current HFEPX corpus.

Generated Apr 10, 2026, 7:25 AM · Grounded in abstract + metadata only

Key Takeaways

  • To study this, we introduce the Graded Color Attribution (GCA) dataset, a controlled benchmark designed to elicit decision rules and evaluate participant faithfulness to these…
  • Using GCA, both VLMs and human participants establish a threshold: the minimum percentage of pixels of a given color an object must have to receive that color label.

Researcher Actions

  • Treat this as method context, then pivot to protocol-specific HFEPX hubs.
  • Identify benchmark choices from full text before operationalizing conclusions.
  • Validate metric comparability (faithfulness).

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

  • To study this, we introduce the Graded Color Attribution (GCA) dataset, a controlled benchmark designed to elicit decision rules and evaluate participant faithfulness to these rules.
  • Using GCA, both VLMs and human participants establish a threshold: the minimum percentage of pixels of a given color an object must have to receive that color label.
  • Human participants remain faithful to their stated rules, with any apparent violations being explained by a well-documented tendency to overestimate color coverage.

Why It Matters For Eval

  • To study this, we introduce the Graded Color Attribution (GCA) dataset, a controlled benchmark designed to elicit decision rules and evaluate participant faithfulness to these rules.
  • Using GCA, both VLMs and human participants establish a threshold: the minimum percentage of pixels of a given color an object must have to receive that color label.

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.

  • Gap: Benchmark or dataset anchors are present

    No benchmark/dataset anchor extracted from abstract.

  • Pass: Metric reporting is present

    Detected: faithfulness

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