Modality Fault Lines: Structural Corruptions Reveal Fragile Omni-Modal Reasoning

• URL: http://arxiv.org/abs/2608.29278v1
• Date: Sat, 29 Aug 2026 14:05:42 GMT
• Title: Modality Fault Lines: Structural Corruptions Reveal Fragile Omni-Modal Reasoning
• Abstract summary: We introduce SCEval, a diagnostic evaluation protocol that keeps the question, answer space, and modality channels fixed while applying controlled structural corruptions to text, vision, and audio individually and jointly.<n>Results show that structural corruption lowers clean accuracy, text--vision damage forms the most stable shared fault line, and multi-modal degradation is non-additive rather than a simple function of the number of corrupted modalities.
• Score: 5.307937307387054
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Omni-modal large language models are increasingly evaluated on clean text--vision--audio inputs, where every channel is present, synchronized, and readily interpretable. Such scores are often taken as evidence of robust cross-modal fusion, but clean evaluation cannot tell whether success depends on stable cross-modal structure or on cues sufficient only in intact inputs. To address this gap, we define a modality fault line: a boundary at which model behavior becomes unstable when a modality remains present and human-interpretable, but its internal evidence structure is perturbed. We introduce SCEval (Structure-Corruption Evaluation) a diagnostic evaluation protocol that keeps the question, answer space, and modality channels fixed while applying controlled structural corruptions to text, vision, and audio individually and jointly. Built from $273$ human-verified tri-modal examples from Social-IQ, OmniBench, and VALOR, SCEval evaluates $15$ proprietary and open-source omni-modal systems. The results show that structural corruption lowers clean accuracy, text--vision damage forms the most stable shared fault line, and multi-modal degradation is non-additive rather than a simple function of the number of corrupted modalities. Clean omni-modal accuracy therefore does not establish that a model will remain reliable when cross-modal evidence becomes structurally unreliable.

Related papers

• Conformal Fusion Under Missing Modalities [6.599344783327054]
  We argue that missing-modality robustness and calibrated uncertainty are a single coupled property, and introduce Modality-Conditioned Conformal Fusion (MCCF)<n>MCCF combines a multimodal bottleneck fusion backbone trained with modality dropout, per-modality evidential heads producing modality-decomposed Dirichlet distributions, and a Dempster-Shafer combination rule.<n>A Mondrian conformal calibration module keyed on the modality-presence mask then provides finite-sample group-conditional coverage for every non-empty modality subset.
  arXiv  Detail & Related papers  (2026-08-07T12:57:16Z)
• C$^3$PO: Evaluating Cross-Modal Composition and Counterfactual Performance in Omnimodal Models [11.992819604307998]
  We introduce C$3$PO, a benchmark of 3,404 samples spanning video, audio, image, and text.<n>We evaluate two abilities: information composition (fusing dispersed evidence) and counterfactual conflict.<n>Built through a fully automatic pipeline using 25 logically grounded templates, C$3$PO reveals that while humans achieve 88.64% accuracy, the best model reaches only 73.17%.
  arXiv  Detail & Related papers  (2026-08-05T20:04:05Z)
• Adaptive Modality Reliability Diagnosis and Restoration for Robust Multimodal Intent Recognition [2.2012643583422347]
  Multimodal intent recognition combines linguistic, acoustic, and visual evidence, but individual modalities may be noisy, missing, semantically conflicting, or disproportionately dominant.<n>We propose PRIME, a closed-loop reliability guided framework that jointly diagnoses, restores, and reassesses modality quality at the sample level.<n> Experiments on multimodal intent-recognition benchmarks show that PRIME maintains competitive clean-data performance while improving under missing, noisy, conflicting, and modality-imbalanced conditions.
  arXiv  Detail & Related papers  (2026-08-04T11:14:13Z)
• Omni-Modal Dissonance Benchmark: Systematically Breaking Modality Consensus to Probe Robustness and Calibrated Abstention [7.218987720005679]
  We introduce OMD-Bench, where all modalities are initially congruent.<n>We then systematically corrupt to isolate each modality's contribution.<n>The benchmark comprises 4,080 instances spanning 27 anchors across eight corruption conditions.
  arXiv  Detail & Related papers  (2026-03-28T08:29:15Z)
• 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)
• Illusions of Confidence? Diagnosing LLM Truthfulness via Neighborhood Consistency [78.91846841708586]
  We show that even facts answered with perfect self-consistency can rapidly collapse under mild contextual interference.<n>We propose Neighbor-Consistency Belief (NCB), a structural measure of belief that evaluates response coherence across a conceptual neighborhood.<n>We also present Structure-Aware Training (SAT), which optimize context-invariant belief structure and reduces long-tail knowledge brittleness by approximately 30%.
  arXiv  Detail & Related papers  (2026-01-09T16:23:21Z)
• All That Glisters Is Not Gold: A Benchmark for Reference-Free Counterfactual Financial Misinformation Detection [67.89888669159899]
  RFC Bench is a benchmark for evaluating large language models on financial misinformation under realistic news.<n>The benchmark defines two complementary tasks: reference free misinformation detection and comparison based diagnosis.
  arXiv  Detail & Related papers  (2026-01-07T18:18:28Z)
• Benchmarking Corruption Robustness of LVLMs: A Discriminative Benchmark and Robustness Alignment Metric [49.393713730706445]
  We introduce Bench-C, a benchmark emphasizing discriminative samples for assessing corruption robustness.<n>We propose the Robustness Alignment Score (RAS), a unified metric that measures degradation in logit-level prediction structure.
  arXiv  Detail & Related papers  (2025-11-24T12:07:56Z)
• SVeritas: Benchmark for Robust Speaker Verification under Diverse Conditions [54.34001921326444]
  Speaker verification (SV) models are increasingly integrated into security, personalization, and access control systems.<n>Existing benchmarks evaluate only subsets of these conditions, missing others entirely.<n>We introduce SVeritas, a comprehensive Speaker Verification tasks benchmark suite, assessing SV systems under stressors like recording duration, spontaneity, content, noise, microphone distance, reverberation, channel mismatches, audio bandwidth, codecs, speaker age, and susceptibility to spoofing and adversarial attacks.
  arXiv  Detail & Related papers  (2025-09-21T14:11:16Z)
• Robustness and Accuracy Could Be Reconcilable by (Proper) Definition [109.62614226793833]
  The trade-off between robustness and accuracy has been widely studied in the adversarial literature. We find that it may stem from the improperly defined robust error, which imposes an inductive bias of local invariance. By definition, SCORE facilitates the reconciliation between robustness and accuracy, while still handling the worst-case uncertainty.
  arXiv  Detail & Related papers  (2022-02-21T10:36:09Z)

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.