AGAR uses middle-to-late layer attention in VLMs to identify and enlarge important word spans in rendered text images, improving performance on visual text comprehension benchmarks.
Zhining Liu, Rana Ali Amjad, Ravinarayana Adkathimar, Tianxin Wei, and Hanghang Tong
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
representative citing papers
Attribution graphs reveal that RAG failures arise from shallow fragmented evidence flow in LLMs, enabling topology-based detection and targeted interventions that reinforce question-guided routing.
HiDe uses token-wise attention decoupling and layout-preserving decoupling to build compact crops that push Qwen2.5-VL and InternVL3 to state-of-the-art scores on high-resolution VQA benchmarks.
citing papers explorer
-
Magnifying What Matters: Attention-Guided Adaptive Rendering for Visual Text Comprehension
AGAR uses middle-to-late layer attention in VLMs to identify and enlarge important word spans in rendered text images, improving performance on visual text comprehension benchmarks.
-
Why Retrieval-Augmented Generation Fails: A Graph Perspective
Attribution graphs reveal that RAG failures arise from shallow fragmented evidence flow in LLMs, enabling topology-based detection and targeted interventions that reinforce question-guided routing.
-
HiDe: Rethinking The Zoom-IN method in High Resolution MLLMs via Hierarchical Decoupling
HiDe uses token-wise attention decoupling and layout-preserving decoupling to build compact crops that push Qwen2.5-VL and InternVL3 to state-of-the-art scores on high-resolution VQA benchmarks.