Omni-Modal Dissonance Benchmark: Systematically Breaking Modality Consensus to Probe Robustness and Calibrated Abstention

TL;DR

OMD-Bench is a systematic benchmark designed to evaluate the robustness, modality reliance, and calibrated abstention of omni-modal models by systematically corrupting modalities and analyzing model responses.

Contribution

The paper introduces OMD-Bench, a novel benchmark that isolates each modality's contribution and assesses models' confidence and abstention behavior under corruption.

Findings

Models over-abstain when two modalities are corrupted.

Models severely under-abstain when all three modalities are corrupted.

Chain-of-thought prompting increases confidence but amplifies overconfidence.

Abstract

Existing omni-modal benchmarks attempt to measure modality-specific contributions, but their measurements are confounded: naturally co-occurring modalities carry correlated yet unequal information, making it unclear whether results reflect true modality reliance or information asymmetry. We introduce OMD-Bench, where all modalities are initially congruent - each presenting the same anchor, an object or event independently perceivable through video, audio, and text - which we then systematically corrupt to isolate each modality's contribution. We also evaluate calibrated abstention: whether models appropriately refrain from answering when evidence is conflicting. The benchmark comprises 4,080 instances spanning 27 anchors across eight corruption conditions. Evaluating ten omni-modal models under zero-shot and chain-of-thought prompting, we find that models over-abstain when two modalities are corrupted yet under-abstain severely when all three are, while maintaining high confidence (~60-100%) even under full corruption. Chain-of-thought prompting improves abstention alignment with human judgment but amplifies overconfidence rather than mitigating it. OMD-Bench provides a diagnostic benchmark for diagnosing modality reliance, robustness to cross-modal inconsistency, and uncertainty calibration in omni-modal systems.