AnswerMap: Faithful Spatial Interpretability of VLMs from Answer Posteriors
cs.CV, cs.LG
Submitted: 2026-09-28
Updated: 2026-09-28
Code: https://github.com/sanataff/AnswerMap
Terminology
Sources
- Qwen3-VL Technical Report
- Explanation-Driven Counterfactual Testing for Faithfulness in Vision-Language Model Explanations
- GridProbe: Posterior-Probing for Adaptive Test-Time Compute in Long-Video VLMs
- LoRA: Low-Rank Adaptation of Large Language Models
- Where Reliability Lives in Vision-Language Models: A Mechanistic Study of Attention, Hidden States, and Causal Circuits
- RISE: Randomized Input Sampling for Explanation of Black-box Models
- GLIMPSE: Holistic Cross-Modal Explainability for Large Vision-Language Models
- Are VLMs Seeing or Just Saying? Uncovering the Illusion of Visual Re-examination
- Visual Attention Faithfulness in Vision-Language Models is Heterogeneous
- InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
- Where do Large Vision-Language Models Look at when Answering Questions?
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models