The paper investigates saliency-guided data augmentation for low-resource handwritten character recognition, proposing that saliency maps can be used to design augmentation variants that perturb, preserve, or constrain visually informative regions. It introduces the Explainability Consistency Index (ECI) to measure saliency stability under perturbations and compares classical transformations, random lacunae, and saliency-constrained lacunae. Results show that saliency-constrained augmentation can improve recognition and explanation consistency, but the optimal strategy depends on architecture and corpus, suggesting saliency should be used as a control mechanism rather than a fixed augmentation rule.
arXiv:2610.28051v1 Announce Type: cross Abstract: Understanding how supervision signals shape semantic visual representations remains challenging, particularly in fine-grained, low-data settings. Sta...
arXiv:2607. 21353v1 Announce Type: new Abstract: Machine unlearning aims to remove the influence of specific training data while preserving model utility.
Understanding how supervision signals shape semantic visual representations remains challenging, particularly in fine-grained, low-data settings. Standard evaluations of vision models rely on downstre...
Machine unlearning aims to remove the influence of specific training data while preserving model utility. Many state-of-the-art approaches pursue this goal by restricting the forgetting update to a subset of parameters selected through gradient-based saliency.
arXiv:2609.10224v1 Announce Type: new Abstract: Vision-language models such as CLIP embed images and text in a shared space, where modality-specific distributions often remain separated. Existing acc...
The paper introduces Target Saliency Boosting, a new task that enhances the visual prominence of a specific object in text-to-image generation without visual priors. It proposes GazeME, a lightweight framework that inserts learnable marker tokens around object descriptions to indicate which objects to emphasize or suppress. By building a saliency-semantics dataset and using Saliency Prior Marker Activation, GazeME learns to adjust markers during training and automatically applies them at inference, effectively boosting target saliency while maintaining semantic alignment and image quality.
arXiv:2609.37581v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) excel at visual understanding and reasoning but often incur substantial inference costs due to the large number of visu...
The paper introduces RoSA, a saliency-aware fine‑tuning framework for long‑text image‑text alignment that addresses the failure mode of saliency collapse. RoSA decomposes image‑text pairs into salient and contextual views using a modality‑independent strategy that combines visual saliency estimation with LLM‑based text decomposition, and employs auxiliary objectives to align these views. The method improves long‑text retrieval performance and robustness to saliency collapse, and its auxiliary objectives can be added to existing tuning methods for consistent gains.
arXiv:2609.36916v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) incur high inference latency from long visual token sequences. Existing pruning methods commonly use attention...
arXiv:2605. 16409v3 Announce Type: replace-cross Abstract: Optical character recognition (OCR) and multilingual scene-text understanding remain challenging for multimodal large language models (MLLMs), particularly in real-world images containing small or degraded text, cluttered layouts, occlusion, handwriting, and complex typography.
The paper introduces a controlled benchmark for evaluating large language models (LLMs) on key‑value pair extraction from documents with varying levels of OCR noise. It tests 136 configurations across five instruction‑tuned open‑weight LLMs, three datasets, and four text‑quality conditions, using deterministic decoding to generate 17,688 document‑level inferences. The study finds that clean‑text performance does not reliably predict real‑world robustness, model rankings can reverse under noisy conditions, and few‑shot demonstrations do not always improve accuracy, highlighting reliability risks in OCR‑to‑LLM pipelines.
arXiv:2607. 24017v1 Announce Type: cross Abstract: The empirical success of attention mechanism in Multimodal Large Language Models (MLLMs) often obscures its inherent, subtle flaws.