# AnswerMap: Faithful Spatial Interpretability of VLMs from Answer Posteriors

arXiv: https://arxiv.org/abs/2609.35247 · HTML: https://mlsanity.com/paper/2609.35247/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 31 of its 32 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## Builds on (most cited first, 30 of 31)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [Visualizing and Understanding Convolutional Networks](https://mlsanity.com/paper/1311.2901/) (2013)
- [Evaluating Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2305.10355/) (2023)
- [Towards VQA Models That Can Read](https://mlsanity.com/paper/1904.08920/) (2019)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Modeling Context in Referring Expressions](https://mlsanity.com/paper/1608.00272/) (2016)
- [Attention is not Explanation](https://mlsanity.com/paper/1902.10186/) (2019)
- [Generation and Comprehension of Unambiguous Object Descriptions](https://mlsanity.com/paper/1511.02283/) (2015)
- [RISE: Randomized Input Sampling for Explanation of Black-box Models](https://mlsanity.com/paper/1806.07421/) (2018)
- [Quantifying Attention Flow in Transformers](https://mlsanity.com/paper/2005.00928/) (2020)
- [An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models](https://mlsanity.com/paper/2403.06764/) (2024)
- [Is Attention Interpretable?](https://mlsanity.com/paper/1906.03731/) (2019)
- [Generic Attention-model Explainability for Interpreting Bi-Modal and Encoder-Decoder Transformers](https://mlsanity.com/paper/2103.15679/) (2021)
- [Lingshu: A Generalist Foundation Model for Unified Multimodal Medical Understanding and Reasoning](https://mlsanity.com/paper/2506.07044/) (2025)
- [What do different evaluation metrics tell us about saliency models?](https://mlsanity.com/paper/1604.03605/) (2016)
- [See What You Are Told: Visual Attention Sink in Large Multimodal Models](https://mlsanity.com/paper/2503.03321/) (2025)
- [Your Large Vision-Language Model Only Needs A Few Attention Heads For Visual Grounding](https://mlsanity.com/paper/2503.06287/) (2025)
- [Token Activation Map to Visually Explain Multimodal LLMs](https://mlsanity.com/paper/2506.23270/) (2025)
- [Do Vision & Language Decoders use Images and Text equally? How Self-consistent are their Explanations?](https://mlsanity.com/paper/2404.18624/) (2024)
- [On the Faithfulness of Vision Transformer Explanations](https://mlsanity.com/paper/2404.01415/) (2024)
- [Where do Large Vision-Language Models Look at when Answering Questions?](https://mlsanity.com/paper/2503.13891/) (2025)
- [Are VLMs Seeing or Just Saying? Uncovering the Illusion of Visual Re-examination](https://mlsanity.com/paper/2605.15864/) (2026)
- [Explanation-Driven Counterfactual Testing for Faithfulness in Vision-Language Model Explanations](https://mlsanity.com/paper/2510.00047/) (2025)
- [GLIMPSE: Holistic Cross-Modal Explainability for Large Vision-Language Models](https://mlsanity.com/paper/2506.18985/) (2025)
- [Large Vision-Language Models Get Lost in Attention](https://mlsanity.com/paper/2605.05668/) (2026)
- [GridProbe: Posterior-Probing for Adaptive Test-Time Compute in Long-Video VLMs](https://mlsanity.com/paper/2605.10762/) (2026)
- [CAVE: Detecting and Explaining Commonsense Anomalies in Visual Environments](https://mlsanity.com/paper/2510.26006/) (2025)
- [Where Reliability Lives in Vision-Language Models: A Mechanistic Study of Attention, Hidden States, and Causal Circuits](https://mlsanity.com/paper/2605.08200/) (2026)

## Abstract (arXiv)

When a VLM answers a visual query, current interpretability tools rely on text rationales, which use a mismatched modality, or on internal read-outs, which originate too early to reflect the final output and require white-box access to the model. We introduce AnswerMap, a training-free, task-agnostic, black-box visual rationale constructed from the output head. The image is cut into K row and K column bands, each shown alone to the frozen model along with the query in the format of a yes/no relevance question. The outer product of the row and column ``yes'' posteriors gives the query-conditioned spatial map. Crucially, by defining a fixed read-out R (e.g., expectation, maximum) on top of AnswerMap, we can derive continuous outputs like location natively. This bypasses the reliance on discrete text tokens for continuous-output tasks and guarantees an image-dependent answer by construction. However, a rationale can be confabulated, so we validate AnswerMap across four models and three query distributions with two tests: (a) agreement with the model's own generated point and (b) deletion of the map's region. The map lands where the model points (AUC 0.85 against 0.38 for attention), and deleting its region flips 53% of correct answers (against 19% for attention's). Beyond establishing faithfulness, we demonstrate the map's task-agnostic utility through three distinct read-outs: its maximum flags hallucinated objects without generation, its expectation localizes correctly when the model's own pointing fails, and its top-mass region, fed back as a crop, fixes half of the model's wrong answers. AnswerMap thus offers a new lens on VLM interpretability and, through its read-outs, a new output interface for visual tasks beyond text tokens.