Sayan Nag

Sayan is a Research Scientist at Adobe Research. His primary research interests broadly include Multimodal Learning (Vision-Language, Audio-Visual), Self-Supervised Learning, and Time-Series Analysis. Prior to joining Adobe, Sayan earned his Ph.D. degree from the University of Toronto, Canada in 2024.

More information, including a list of publications, can be found here.

AI & Machine Learning Computer Vision, Imaging & Video Intelligent Agents & Assistants

Publications

September 12, 2026 Workshop on Multimodal Algorithmic Reasoning for Scenes and Structures (MARS2) at IEEE/CVF European Conference on Computer Vision (ECCV)

Reward-Validated Coactive Learning for Stylized Slide Design

Avijit Dasgupta, Joseph K J, Pannaga Shivaswamy, Sunandini Sanyal, Sambaran Bandyopadhyay, Sayan Nag, Balaji Vasan Srinivasan, CV Jawahar, Karteek Alahari

September 12, 2026 Workshop on Multimodal Algorithmic Reasoning for Scenes and Structures (MARS2) at IEEE/CVF European Conference on Computer Vision (ECCV)

Reasoning Traces in MLLMs Offer Rich Guidance for Design Transformation

Aveen Dayan, Joseph K J, Sayan Nag, Balaji Vasan Srinivasan

June 3, 2026 Conference on Computer Vision and Pattern Recognition (CVPR 2026)

SliderEdit: Continuous Image Editing with Fine-Grained Instruction Control

Arman Zarei, Samyadeep Basu, Mobina Pournemat, Sayan Nag, Ryan A. Rossi, Soheil Feizi

CVPR Oral

June 3, 2026 Conference on Computer Vision and Pattern Recognition (CVPR 2026)

Object-WIPER : Training-Free Object and Associated Effect Removal in Videos

Saksham Singh Kushwaha, Sayan Nag, Yapeng Tian, Kuldeep Kulkarni

June 3, 2026 Conference on Computer Vision and Pattern Recognition (CVPR 2026) Findings

Through the PRISM: Principle-Aware, Interpretable, and Multi-Scale Evaluation of Visual Designs

Mona Gandhi, Joseph K J, Srinivasan Parthasarathy, Sayan Nag

Findings Track

January 20, 2026 AAAI Conference on Artificial Intelligence 2026

Agentic Design Review System

Sayan Nag, Joseph K J, Koustava Goswami, Vlad Morariu, Balaji Vasan Srinivasan

December 1, 2025 Neural Information Processing Systems (NeurIPS)

Localizing Knowledge in Diffusion Transformers

Arman Zarei, Samyadeep Basu, Keivan Rezaei, Zihao Lin, Sayan Nag, Soheil Feizi

September 29, 2024 European Conference on Computer Vision (ECCV)

SAFARI: Adaptive Sequence Transformer for Weakly Supervised Referring Expression Segmentation

Sayan Nag, Koustava Goswami, Srikrishna Karanam

June 17, 2024 CVPR 2024

MeLFusion: Synthesizing Music from Image and Language Cues using Diffusion Models

Sanjoy Chowdhury, Sayan Nag, Joseph K J, Balaji Vasan Srinivasan, Dinesh Manocha

CVPR Highlight

News

Giving Image Editors Adjustable Control Over Editing Instructions: SliderEdit at CVPR 2026

SliderEdit introduces continuous, adjustable control over individual editing instructions in instruction-based image editing models, turning each instruction into a smooth slider.

May 27, 2026

Agentic Design Review System: Teaching AI to review graphic designs the way experts do

Our Agentic Design Review System (Agentic-DRS) is an experimental multi-agent framework from Adobe Research for holistic graphic design evaluation. Specialized AI agents assess typography, color harmony, alignment, spacing, composition, and more, coordinated by a meta-agent.

May 6, 2026

Object-WIPER: A Smarter Way to Erase Objects from Video

Rather than treating object removal as a simple mask-and-fill task, Object-WIPER recognizes that objects leave visual traces in a scene — shadows on the ground, reflections in glass, translucent overlaps with the background — and that a complete removal means erasing those traces too.

April 30, 2026