arXiv Machine Learning
Integrating attention into explanation frameworks for language and vision transformers
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv Machine Learning
Oct 8
The Metagame of Interpretability and Meta-Attributions
arXiv:2605.06295v2 Announce Type: replace Abstract: How can an arbitrary attribution method be generalized from first principles to capture interactions? We answer this with the metagame, a conceptua...
arXiv Machine Learning
Jun 8
OPTIMUS-Prime: Minimal and Sufficient Concept Explanations for Deep Vision Models
arXiv:2606. 07180v1 Announce Type: cross Abstract: The growing demand for transparency in automated decision-making has propelled eXplainable Artificial Intelligence (XAI) to the forefront of machine learning research.
arXiv Machine Learning
Aug 4
LAYA: Layer-wise Attention Aggregation for Interpretable Depth-Aware Neural Networks
arXiv:2511. 12723v2 Announce Type: replace Abstract: Deep neural networks typically rely on the representation produced by their final hidden layer to make predictions, implicitly assuming that this single vector fully captures the semantics encoded across all preceding transformations.
arXiv AI
Jul 13
Explaining is Harder Than Predicting Alone: Evaluating Concept-based Explanations of MLLMs as ICL Visual Classifiers
arXiv:2605. 28215v2 Announce Type: replace Abstract: In-context learning (ICL) enables multimodal large language models (MLLMs) to classify images from a few labelled examples.
arXiv Computer Vision
Oct 16
Perception or Reasoning? Diagnosing and Tackling Reinforcement Learning Bottlenecks in VLMs
The paper introduces Perception and Reasoning Attribution (PARA), a method that quantifies how much of a vision‑language model’s errors can be fixed by better perception versus stronger reasoning. PARA reveals that reinforcement learning (RL) benefits more from improving reasoning than perception, and motivates a new group‑relative attention distillation objective called GRAD. GRAD directly guides weaker rollouts toward the visual attention patterns of the best rollouts, consistently boosting performance across multiple model scales and benchmarks.
arXiv AI
Sep 2
Visual Attention Faithfulness in Vision-Language Models is Heterogeneous
The study investigates whether attention weights in Vision‑Language Models (VLMs) accurately reflect model reasoning for visual inputs. Using causal perturbation analysis, it identifies three distinct processing modes—Faithful‑Sufficient, Faithful‑Distributed, and Non‑Focal—indicating heterogeneous visual attention faithfulness. The research also shows that human‑annotated ground‑truth regions align with model attention in only about 60% of cases, highlighting a systematic divergence between model visual reliance and human intuition across VQA, document, and chart tasks.