DocPrune is a training-free token pruning method that removes background and irrelevant tokens from document images using question and comprehension signals, yielding 3x encoder and 3.3x decoder throughput gains plus +1 F1 on M3DocRAG.
arXiv:2504.04974 [cs.CV] https://arxiv.org/abs/2504.04974
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5representative citing papers
A model-agnostic Geometric Risk Controller reduces extreme errors in VLM-based OCR by requiring cross-view consensus before accepting outputs.
ReAlign improves visual document retrieval by training retrievers to match query-induced rankings with rankings derived from VLM-generated, region-focused descriptions of relevant page content.
PreciseDoc is an LMM that uses synthetic document data pipelines and RL-based joint grounding-reasoning training to achieve higher precision in locating elements within text-rich document images than prior approaches.
citing papers explorer
-
DocPrune:Efficient Document Question Answering via Background, Question, and Comprehension-aware Token Pruning
DocPrune is a training-free token pruning method that removes background and irrelevant tokens from document images using question and comprehension signals, yielding 3x encoder and 3.3x decoder throughput gains plus +1 F1 on M3DocRAG.
-
From Plausibility to Verifiability: Risk-Controlled Generative OCR with Vision-Language Models
A model-agnostic Geometric Risk Controller reduces extreme errors in VLM-based OCR by requiring cross-view consensus before accepting outputs.
-
ReAlign: Optimizing the Visual Document Retriever with Reasoning-Guided Fine-Grained Alignment
ReAlign improves visual document retrieval by training retrievers to match query-induced rankings with rankings derived from VLM-generated, region-focused descriptions of relevant page content.
-
An LMM for Precisely Grounding Elements in Documents
PreciseDoc is an LMM that uses synthetic document data pipelines and RL-based joint grounding-reasoning training to achieve higher precision in locating elements within text-rich document images than prior approaches.
- Reasoning Dynamics and the Limits of Monitoring Modality Reliance in Vision-Language Models