Edge Reliability Gap in Vision-Language Models: Quantifying Failure Modes of Compressed VLMs Under Visual Corruption

• URL: http://arxiv.org/abs/2603.26769v1
• Date: Tue, 24 Mar 2026 10:14:40 GMT
• Title: Edge Reliability Gap in Vision-Language Models: Quantifying Failure Modes of Compressed VLMs Under Visual Corruption
• Abstract summary: The rapid compression of large vision-language models for edge deployment raises an underexplored question: do compact models fail differently, not merely more often?<n>This study compares a 7-billion- parameter quantised VLM (Qwen2.5-VL-7B, 4-bit NF4) against a 500-million- parameter FP16 model (SmolVLM2-500M) across 4,000 samples from VQAv2 and COCO Captions.<n>A three-category error taxonomy (Object Blindness, Semantic Drift, Prior Bias) is applied as a diagnostic framework. Confidence calibration is measured via Expected Error (ECE) using mean token probability
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The rapid compression of large vision-language models (VLMs) for edge deployment raises an underexplored question: do compact models fail differently, not merely more often? This study compares a 7-billion-parameter quantised VLM (Qwen2.5-VL-7B, 4-bit NF4) against a 500-million-parameter FP16 model (SmolVLM2-500M) across 4,000 samples from VQAv2 and COCO Captions. A three-category error taxonomy (Object Blindness, Semantic Drift, Prior Bias) is applied as a diagnostic framework. A text-only GPT-4o judge reveals Semantic Drift (B) as the dominant failure mode on VQAv2 and on COCO for Qwen, with a mixed Object Blindness / Semantic Drift profile for SmolVLM2 on COCO; Prior Bias (C) is present on VQAv2 but absent on COCO for both models. Confidence calibration is measured via Expected Calibration Error (ECE) using geometric mean token probability, compositional reasoning is probed with structured negation probes across four templates, and a blur robustness experiment completes the evaluation. For this model pair, the compact model exhibits a qualitatively distinct failure signature: a 12.5pp larger negation collapse (-33.2pp vs. -20.8pp, Wald 95% CI [8.2, 16.8]pp, p < 10^-8), driven almost entirely by COCO while the VQAv2 gap is not statistically significant (4.5pp, p=0.19). The most discriminating template is false_yn: SMOLVLM2-500M responds "Yes" (incorrectly claiming a depicted object is absent) on 100% of COCO trials vs. 14% for Q WEN 2.5-VL-7B. Asymmetric dataset-dependent miscalibration and a blur experiment with two controlled ablations complete the analysis. The fully reproducible pipeline is released for systematic safety auditing of compressed VLMs prior to edge deployment.

Related papers

• Silent Failures in Quantized LLM Reasoning: A Taxonomy-Based Analysis of Hollow Convergence and Failure Mode Shifts [0.0]
  We show that post-training quantization can silently alter how large language models reason even when task accuracy is preserved.<n>We classify 30,000 chain-of-thought outputs from five instruction-tuned LLMs across three quantization precisions.<n>We find that while accuracy is robust across precisions, Hollow Convergence shows a significant size-dependent shift under NF4.
  arXiv  Detail & Related papers  (2026-07-10T21:55:05Z)
• Modeling Normal Is All You Need: Joint Latent Clustering for Anomaly Detection in Multimodal Cyber-Physical Systems [47.027290803102666]
  Normal behaviour is the union of many imbalanced, curved, thin-fringed operating regimes.<n>We model the normal law with a jointly learned latent representation plus explicit Gaussian-mixture mode clustering.<n>We evaluate under a deliberately fair protocol: raw point-wise metrics with no point adjustment, a trivial-detector difficulty split, prevalence-matched F1, and train-normal-only calibration.
  arXiv  Detail & Related papers  (2026-07-07T10:10:00Z)
• A Controlled Audit of Pretraining Contamination in Public Medical Vision-Language Benchmarks [46.43161388634735]
  Medical vision-language models (VLMs) are evaluated on public benchmarks whose images and question-answer pairs have been freely downloadable for years.<n>We audit open VLMs on SLAKE-En, PathVQA, VQA-RAD, and an auxiliary public OmniMedVQA mirror using four detector families.
  arXiv  Detail & Related papers  (2026-06-08T18:40:10Z)
• Measuring Five-Nines Reliability: Sample-Efficient LLM Evaluation in Saturated Benchmarks [45.86413490112477]
  Large language models (LLMs) are used in reliability-sensitive applications.<n>Estimateing rare failure probability with tight confidence bounds requires prohibitively large LLM inference sizes.<n>We propose to learn a sampling distribution concentrated on failure-prone inputs via the cross-entropy method.
  arXiv  Detail & Related papers  (2026-05-11T20:23:44Z)
• The Metacognitive Probe: Five Behavioural Calibration Diagnostics for LLMs [0.0]
  The Metacognitive Probe is an exploratory five-task, 15-slot diagnostic.<n>It decomposes an LLM's confidence behaviour into five behaviourally-distinct dimensions.<n>It is evaluated on N=69 humans.
  arXiv  Detail & Related papers  (2026-05-11T00:55:31Z)
• Quantization Undoes Alignment: Bias Emergence in Compressed LLMs Across Models and Precision Levels [0.0]
  Large Language Models are routinely compressed via post-training quantization to reduce inference costs and memory footprint for cloud and edge deployment.<n>Existing studies typically compare only two conditions, rely on aggregate bias metrics, and evaluate a single model family.<n>We conduct a controlled empirical study of three instruction-tuned models at five precision levels (BF16 through 3-bit) on 12,148 BBQ bias benchmark items.
  arXiv  Detail & Related papers  (2026-05-02T05:41:47Z)
• Auditing Frontier Vision-Language Models for Trustworthy Medical VQA: Grounding Failures, Format Collapse, and Domain Adaptation [4.177028541321909]
  We audit five vision-language models (VLMs) on Medical VQA along two trust-relevant axes.<n>Models localize anatomical and pathological targets poorly.<n>A self-grounding pipeline, where the same model localizes then answers, degrades VQA accuracy.
  arXiv  Detail & Related papers  (2026-04-30T11:11:47Z)
• The Verification Tax: Fundamental Limits of AI Auditing in the Rare-Error Regime [0.0]
  Most cited calibration result in deep learning -- post-temperature-scaling ECE of 0.012 on CIFAR-100 -- is below the statistical noise floor.<n>We prove this is not a failure of the experiment but a law: the minimax rate for estimating calibration error with model error rate epsilon is Theta((Lepsilon/m)2/3), and no estimator can beat it.
  arXiv  Detail & Related papers  (2026-04-14T16:48:24Z)
• How Independent are Large Language Models? A Statistical Framework for Auditing Behavioral Entanglement and Reweighting Verifier Ensembles [46.63622714488747]
  Shared pretraining data, distillation, and alignment pipelines can induce hidden behavioral dependencies, latent entanglement.<n>In practice, this manifests as correlated reasoning patterns and synchronized failures.<n>We develop a statistical framework for auditing behavioral entanglement among black-box language models.
  arXiv  Detail & Related papers  (2026-04-08T23:32:06Z)
• The Alignment Tax: Response Homogenization in Aligned LLMs and Its Implications for Uncertainty Estimation [1.8345614451086532]
  RLHF-aligned language models exhibit response homogenization on TruthfulQA.<n>40-79% of questions produce a single semantic cluster across 10 i.i.d. samples.
  arXiv  Detail & Related papers  (2026-03-25T09:35:15Z)
• VLM-RobustBench: A Comprehensive Benchmark for Robustness of Vision-Language Models [64.56065206447788]
  Vision-language models (VLMs) achieve strong performance on standard, high-quality datasets.<n>We present VLM-RobustBench, a benchmark spanning 49 augmentation types across noise, blur, weather, digital, and geometrics.<n>We show that low-severity spatial perturbations often degrade performance more than visually severe photometric corruptions.
  arXiv  Detail & Related papers  (2026-03-06T10:58:02Z)
• The Devil in the Details: Emergent Misalignment, Format and Coherence in Open-Weights LLMs [0.0]
  We evaluate if current-generation open-weights models exhibit similar resistance to the Qwen-2.5 family.<n>We replicate the effect across nine modern open-weights models.
  arXiv  Detail & Related papers  (2025-11-25T09:25:33Z)
• Parrot: Persuasion and Agreement Robustness Rating of Output Truth -- A Sycophancy Robustness Benchmark for LLMs [0.0]
  PARROT (Persuasion and Agreement Robustness Rating of Output Truth) is a robustness focused framework designed to measure the degradation in accuracy under social pressure exerted on users.<n>We evaluate 22 models using 1,302 MMLU-style multiple-choice questions across 13 domains and domain-specific authority templates.
  arXiv  Detail & Related papers  (2025-11-21T13:01:28Z)
• Token-Level Inference-Time Alignment for Vision-Language Models [58.41370989069588]
  Vision-Language Models (VLMs) have become essential backbones of modern multimodal intelligence.<n>We present TITA, a lightweight framework that freezes the base VLM and instead trains a reward model to approximate its distribution.<n>During inference, implicit preference signals are extracted as log-probability ratios between the reward model and the target VLM, yielding dense autoregressive feedback.
  arXiv  Detail & Related papers  (2025-10-20T09:58:03Z)
• SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvement [100.85923086072204]
  We introduce ThinkLite-VL, a family of visual reasoning models that achieve state-of-the-art (SoTA) performance using an order of magnitude fewer training samples.<n>We use Monte Carlo Tree Search (MCTS) to measure sample difficulty via the number of reasoning iterations a vision-language model (VLM) requires to solve each instance.<n>ThinkLite-VL-7B and ThinkLite-VL-72B significantly outperform their respective base models across eight visual reasoning benchmarks.
  arXiv  Detail & Related papers  (2025-04-10T17:49:05Z)
• Evaluating Large Language Models for Automated Clinical Abstraction in Pulmonary Embolism Registries: Performance Across Model Sizes, Versions, and Parameters [16.74673750576054]
  We evaluated whether openly available large-language models (LLMs) can automate concept extraction from computed-tomography PE (CTPE) reports without sacrificing data quality.<n>LLMs offer a scalable, accurate solution for PE registry abstraction, and a dual-model review workflow can further safeguard data quality with minimal human oversight.
  arXiv  Detail & Related papers  (2025-03-26T21:38:06Z)
• RLAIF-V: Open-Source AI Feedback Leads to Super GPT-4V Trustworthiness [88.82621231987815]
  We introduce RLAIF-V, a novel framework that aligns MLLMs in a fully open-source paradigm.<n> RLAIF-V maximally explores open-source MLLMs from two perspectives, including high-quality feedback data generation.<n>Experiments on six benchmarks in both automatic and human evaluation show that RLAIF-V substantially enhances the trustworthiness of models.
  arXiv  Detail & Related papers  (2024-05-27T14:37:01Z)
• EqCo: Equivalent Rules for Self-supervised Contrastive Learning [81.45848885547754]
  We propose a method to make self-supervised learning irrelevant to the number of negative samples in InfoNCE-based contrastive learning frameworks. Inspired by the InfoMax principle, we point that the margin term in contrastive loss needs to be adaptively scaled according to the number of negative pairs.
  arXiv  Detail & Related papers  (2020-10-05T11:39:04Z)

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.