# A Low Grounding Score Is Not an Ungrounded Judge: Identifying the Perceptibility Confound in Multimodal Oversight

arXiv: https://arxiv.org/abs/2610.00111 · HTML: https://mlsanity.com/paper/2610.00111/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 36 of its 39 references that are in the map.
- First posted on arXiv 9 Sep 2026.

## Builds on (most cited first, 30 of 36)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts](https://mlsanity.com/paper/2310.02255/) (2023)
- [ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning](https://mlsanity.com/paper/2203.10244/) (2022)
- [A Survey on LLM-as-a-Judge](https://mlsanity.com/paper/2411.15594/) (2024)
- [Scaling Laws for Reward Model Overoptimization](https://mlsanity.com/paper/2210.10760/) (2022)
- [HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Models](https://mlsanity.com/paper/2310.14566/) (2023)
- [MLLM-as-a-Judge: Assessing Multimodal LLM-as-a-Judge with Vision-Language Benchmark](https://mlsanity.com/paper/2402.04788/) (2024)
- [FigureQA: An Annotated Figure Dataset for Visual Reasoning](https://mlsanity.com/paper/1710.07300/) (2017)
- [VisualPRM: An Effective Process Reward Model for Multimodal Reasoning](https://mlsanity.com/paper/2503.10291/) (2025)
- [On scalable oversight with weak LLMs judging strong LLMs](https://mlsanity.com/paper/2407.04622/) (2024)
- [VL-RewardBench: A Challenging Benchmark for Vision-Language Generative Reward Models](https://mlsanity.com/paper/2411.17451/) (2024)
- [Vision Language Models are Biased](https://mlsanity.com/paper/2505.23941/) (2025)
- [Multimodal RewardBench: Holistic Evaluation of Reward Models for Vision Language Models](https://mlsanity.com/paper/2502.14191/) (2025)
- [Treble Counterfactual VLMs: A Causal Approach to Hallucination](https://mlsanity.com/paper/2503.06169/) (2025)
- [Looking Beyond Text: Reducing Language bias in Large Vision-Language Models via Multimodal Dual-Attention and Soft-Image Guidance](https://mlsanity.com/paper/2411.14279/) (2024)
- [Multimodal RewardBench 2: Evaluating Omni Reward Models for Interleaved Text and Image](https://mlsanity.com/paper/2512.16899/) (2025)
- [Are VLMs Seeing or Just Saying? Uncovering the Illusion of Visual Re-examination](https://mlsanity.com/paper/2605.15864/) (2026)
- [VRPRM: Process Reward Modeling via Visual Reasoning](https://mlsanity.com/paper/2508.03556/) (2025)
- [Arbitration Failure, Not Perceptual Blindness: How Vision-Language Models Resolve Visual-Linguistic Conflicts](https://mlsanity.com/paper/2604.09364/) (2026)
- [Fooling the LVLM Judges: Visual Biases in LVLM-Based Evaluation](https://mlsanity.com/paper/2505.15249/) (2025)
- [Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models](https://mlsanity.com/paper/2603.06621/) (2026)
- [Reward Auditor: Inference on Reward Modeling Suitability in Real-World Perturbed Scenarios](https://mlsanity.com/paper/2512.00920/) (2025)
- [VLM Judges Can Rank but Cannot Score: Task-Dependent Uncertainty in Multimodal Evaluation](https://mlsanity.com/paper/2604.25235/) (2026)
- [Do Medical Vision Language Models Actually See? A Counterfactual Grounding Framework and Hard-Negative Contrastive Training for Visually-Reliant Medical VLMs](https://mlsanity.com/paper/2607.03647/) (2026)
- [Faithful GRPO: Improving Visual Spatial Reasoning in Multimodal Language Models via Constrained Policy Optimization](https://mlsanity.com/paper/2604.08476/) (2026)
- [Grounding the Score: Explicit Visual Premise Verification for Reliable Vision-Language Process Reward Models](https://mlsanity.com/paper/2603.16253/) (2026)
- [Improving Alignment in LVLMs with Debiased Self-Judgment](https://mlsanity.com/paper/2508.20655/) (2025)
- [MJ1: Multimodal Judgment via Grounded Verification](https://mlsanity.com/paper/2603.07990/) (2026)

## Abstract (arXiv)

Model judges now supervise multimodal systems at scale, filtering training data, selecting outputs, and supplying the reward that shapes multimodal reasoning models. Trusting one means first checking that it uses its evidence, and that check is itself worth scrutinizing, so we ask whether a counterfactual probe of visual grounding measures what it claims to. The probe edits the image so the ground truth flips, holds the reasoning trace fixed, and asks whether the verdict follows. We formalize it as the Verdict Grounding Score and show it cannot be read the way such scores are read. A verdict responds only to an edit that reaches the judge's decision-relevant reading, so the score is capped by how perceptible the edit is, and unless editing makes the attribute easier to read, the error is one-sided: the score can only make a judge look less grounded than it is. The practical failure is therefore a false alarm, an auditor discarding a usable overseer. Under assumptions we state, we show this missing quantity is not merely bounded but identified from three quantities the same audit protocol already collects, which makes the false-alarm rate directly measurable rather than merely a concern. Auditing nine judges, we find the predicted ordering holds strictly across our entire primary pool, and the typical judge there acts on only about half of the edits whose attribute it can otherwise resolve. Applying a conservative rejection threshold certifies several cells as false alarms outright, the clearest being a judge that detects the injected error essentially every time while still scoring as if it had not used the image at all. The rule that follows is that an image-side counterfactual score should never be reported alone: a detection probe on the unedited image upper-bounds it, certifies its false alarms, and costs nothing extra to run.