Small Vision-Language Models Know When They Are Wrong But Cannot Say So: A Two-Model Study of Stated versus Internal Confidence Under Realistic Image Degradation

• URL: http://arxiv.org/abs/2607.22034v1
• Date: Fri, 24 Jul 2026 07:00:32 GMT
• Title: Small Vision-Language Models Know When They Are Wrong But Cannot Say So: A Two-Model Study of Stated versus Internal Confidence Under Realistic Image Degradation
• Abstract summary: Vision-language models (VLMs) are increasingly deployed on consumer hardware where input images are degraded by compression, camera shake, and poor lighting.<n>We evaluate two small open-weight VLMs -- Qwen2-VL-2B-Instruct and SmolVLM-Instruct -- across six realistic photographic degradations at three severity levels.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-language models (VLMs) are increasingly deployed on consumer hardware where input images are degraded by compression, camera shake, and poor lighting. In such settings, a reliable uncertainty signal matters more than raw accuracy, because it determines when a system should defer rather than answer. We evaluate two small open-weight VLMs -- Qwen2-VL-2B-Instruct and SmolVLM-Instruct -- across six realistic photographic degradations at three severity levels, comparing two confidence signals: the confidence the model states in natural language, and the model's own mean token probability over its generated answer. Across 3,800 predictions, we find a large and consistent gap. Verbalized confidence in Qwen2-VL is almost constant (mean 0.87-0.90 across all conditions) and detects its own errors at chance level (AUROC 0.39-0.75, typically ~0.50), while internal token probability from the same model separates correct from incorrect answers with AUROC 0.92-0.99. In SmolVLM, verbalized confidence proved largely unobtainable: across three prompt templates, only one of five pilot attempts produced a parseable confidence value, while internal probability again yielded above-chance error detection (AUROC 0.54-0.92). Both models fail in the same place: under severe underexposure, accuracy collapses (0.99->0.22 for Qwen2-VL, 0.97->0.42 for SmolVLM) while both confidence signals barely move, and internal error-detection falls to chance. We conclude that small VLMs encode usable self-knowledge that their verbalized output does not express, that internal probability is therefore the better deferral signal in constrained deployment, and that neither signal should be trusted under severe low-light conditions.

Related papers

• HalluPrism: When Multimodal Uncertainty Should Diagnose, Not Decide [18.50218795316691]
  We propose HalluPrism, a behavioral diagnostic that re-runs an answer after visual degradation, blank-image replacement, and grounding or relation checks.<n>These targeted probes yield a signature over visual-perturbation sensitivity, image-removal confidence retention, and grounding/relation-probe instability.
  arXiv  Detail & Related papers  (2026-08-29T11:06:37Z)
• Accurate Ensembles, Fragile Narratives: Multi-Scale Stacking and a Fidelity Audit of LLM-Generated Explanations for Credit Risk [0.0]
  Credit scoring increasingly relies on models whose decision logic cannot be read off their parameters.<n>A common proposal closes that gap with a language model: compute feature attributions, hand them to an LLM, and let it write the rationale.<n>We build such a system end to end and test whether the second half of the promise holds.
  arXiv  Detail & Related papers  (2026-08-08T13:22:14Z)
• Bigger or Cheaper? Scale and Quantization Effects on Uncertainty Signals in Vision-Language Models Under Image Degradation [0.0]
  We measure, on identical inputs, how model scale and 4-bit quantization affect two confidence signals in the Qwen2-VL family.<n>We find that scale sharply improves the model's internal uncertainty signal.<n>We argue that error-detection AUROC, not calibration error, is the metric that exposes the difference between the two signals.
  arXiv  Detail & Related papers  (2026-07-27T13:47:44Z)
• Speaking in Self-Assessing Tongues: On the Verbalized Confidence of LLMs in Machine Translation [20.39535344813245]
  Unsupervised approaches based on internal signals like predicted probabilities can be misleading because they reflect certainty among alternatives rather than correctness.<n>We devise five verbalized methods of extracting an LLM's per-token confidence without those shortcomings and compare their reliability with that of the model's internal signals of certainty.<n>We find little to no correlation between internal and verbalized methods.
  arXiv  Detail & Related papers  (2026-06-15T19:27:47Z)
• CaliDist: Calibrating Large Language Models via Behavioral Robustness to Distraction [51.56484100374058]
  Existing calibration methods for Large Language Models (LLMs) often overlook a critical dimension of trustworthiness: a model's behavioral robustness.<n>We introduce textscCaliDist, a novel post-hoc calibration approach that directly measures and penalizes a model's susceptibility to distraction.<n>textscCaliDist consistently achieves lower Expected Error (ECE) and Brier Score compared with strong baselines.
  arXiv  Detail & Related papers  (2026-06-04T07:27:53Z)
• Discovery of Hidden Miscalibration Regimes [52.452902154360565]
  Models may be systematically overconfident on some kinds of inputs and underconfident on others.<n>We define the corresponding miscalibration field and propose a diagnostic framework for estimating it.<n>Our approach learns a calibration-aware representation of the input space and estimates signed local miscalibration by kernel smoothing in the learned geometry.
  arXiv  Detail & Related papers  (2026-05-13T13:07:50Z)
• What Single-Prompt Accuracy Misses: A Multi-Variant Reliability Audit of Language Models [0.0]
  Single-prompt accuracy is the dominant way to benchmark language models, but it can miss reliability failures that matter.<n>We evaluate a 15-model open-weight corpus, with the main reliability analyses focused on 10 instruct models across five classification and reasoning benchmarks.<n>We find three broad results. First, evaluation design can materially change the conclusion.<n>Second, confidence signals are fragile. On MMLU-Pro, every primary model verbally reports confidence substantially above both its accuracy and its token-probability confidence on the same rows, and verbal parse rate can collapse for a single model on a single prompt variant.
  arXiv  Detail & Related papers  (2026-05-03T20:05:08Z)
• Verbal Confidence Saturation in 3-9B Open-Weight Instruction-Tuned LLMs: A Pre-Registered Psychometric Validity Screen [0.0]
  We tested whether seven instruction-tuned open-weight models produce verbalised confidence that meets minimal validity criteria.<n>We found Categorical elicitation did not rescue validity.<n> Token-level logprobability did not usefully predict verbalised confidence under the observed variance regime.
  arXiv  Detail & Related papers  (2026-04-24T04:45:21Z)
• Same Answer, Different Representations: Hidden instability in VLMs [65.36933543377346]
  We introduce a representation-aware and frequency-aware evaluation framework that measures internal embedding drift, spectral sensitivity, and structural smoothness.<n>We apply this framework to modern Vision Language Models (VLMs) across the SEEDBench, MMMU, and POPE datasets.
  arXiv  Detail & Related papers  (2026-02-06T12:24:26Z)
• Explicit Abstention Knobs for Predictable Reliability in Video Question Answering [0.41288208387994896]
  High-stakes deployment of vision-language models requires selective prediction.<n>We investigate whether confidence-based abstention provides reliable control over error rates in video question answering.
  arXiv  Detail & Related papers  (2025-12-31T23:27:32Z)
• ConfTuner: Training Large Language Models to Express Their Confidence Verbally [58.63318088243125]
  Large Language Models (LLMs) are increasingly deployed in high-stakes domains such as science, law, and healthcare.<n>LLMs are often observed to generate incorrect answers with high confidence, a phenomenon known as "overconfidence"
  arXiv  Detail & Related papers  (2025-08-26T09:25:32Z)
• Can LLMs Detect Their Confabulations? Estimating Reliability in Uncertainty-Aware Language Models [24.72990207218907]
  Large Language Models (LLMs) are prone to generating fluent but incorrect content, known as confabulation.<n>We investigate how in-context information influences model behavior and whether LLMs can identify their unreliable responses.
  arXiv  Detail & Related papers  (2025-08-11T16:12:36Z)
• Robust Fine-tuning of Zero-shot Models via Variance Reduction [56.360865951192324]
  When fine-tuning zero-shot models, our desideratum is for the fine-tuned model to excel in both in-distribution (ID) and out-of-distribution (OOD) We propose a sample-wise ensembling technique that can simultaneously attain the best ID and OOD accuracy without the trade-offs.
  arXiv  Detail & Related papers  (2024-11-11T13:13:39Z)
• Conservative Prediction via Data-Driven Confidence Minimization [70.93946578046003]
  In safety-critical applications of machine learning, it is often desirable for a model to be conservative. We propose the Data-Driven Confidence Minimization framework, which minimizes confidence on an uncertainty dataset.
  arXiv  Detail & Related papers  (2023-06-08T07:05:36Z)
• Reliability-Aware Prediction via Uncertainty Learning for Person Image Retrieval [51.83967175585896]
  UAL aims at providing reliability-aware predictions by considering data uncertainty and model uncertainty simultaneously. Data uncertainty captures the noise" inherent in the sample, while model uncertainty depicts the model's confidence in the sample's prediction.
  arXiv  Detail & Related papers  (2022-10-24T17:53:20Z)

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.