5 papers
Can Circuit Alignment Predict OOD Generalization?
Ayan Banerjee, Abhra Chaudhuri, Josep Llados +2
Can out-of-distribution (OOD) generalization be predicted from a trained model's weights alone, without any target-domain data? Existing representational similarity metrics (CKA, S…
DocRevive: A Unified Pipeline for Document Text Restoration
Kunal Purkayastha, Ayan Banerjee, Josep Llados +1
In Document Understanding, the challenge of reconstructing damaged, occluded, or incomplete text remains a critical yet unexplored problem. Subsequent document understanding tasks…
TaleDiffusion: Multi-Character Story Generation with Dialogue Rendering
Ayan Banerjee, Josep Llados, Umapada Pal +1
Text-to-story visualization is challenging due to the need for consistent interaction among multiple characters across frames. Existing methods struggle with character consistency,…
CraftGraffiti: Exploring Human Identity with Custom Graffiti Art via Facial-Preserving Diffusion Models
Ayan Banerjee, Fernando Vilariño, Josep Lladós
Preserving facial identity under extreme stylistic transformation remains a major challenge in generative art. In graffiti, a high-contrast, abstract medium, subtle distortions to…
CountLoop: Training-Free High-Instance Image Generation via Iterative Agent Guidance
Anindya Mondal, Ayan Banerjee, Sauradip Nag +3
Diffusion models excel at photorealistic synthesis but struggle with object count fidelity, especially in high-density settings. We introduce COUNTLOOP, a training-free framework t…