# Good Scores, Bad Data: A Metric for Multimodal Coherence

Source: [Good Scores, Bad Data: A Metric for Multimodal Coherence](<https://www.sota2.com/research/paper/69ca039c0636e5bf4afc9418>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-17T05:00:13.532Z

Authors: Vasundra Srinivasan

Published: 2026-03-26T21:30:34.000Z

## Abstract

Multimodal AI systems are evaluated by downstream task accuracy, but high accuracy does not mean the underlying data is coherent. A model can score well on Visual Question Answering (VQA) while its inputs contradict each other. We introduce the Multimodal Coherence Score (MCS), a metric that evaluates fusion quality independent of any downstream model. MCS decomposes coherence into four dimensions, identity, spatial, semantic, and decision, with weights learned via Nelder-Mead optimization. We evaluate on 1,000 Visual Genome images using DETR, CLIP, and ViLT, and validate on 150 COCO images with no retraining. Across three fusion architectures, MCS discriminates quality with higher sensitivity than task accuracy alone (Spearman rho = 0.093 vs. 0.071). Perturbation experiments confirm each dimension responds independently to its failure mode with zero cross-talk. MCS is lightweight, requires no human annotation, and tells you not just that something broke, but what broke.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.25924>)

[PDF](<https://arxiv.org/pdf/2603.25924.pdf>)