Goirik Chakrabarty

My name is Goirik Chakrabarty, and I am Ph.D. candidate in Data Science at the University of Göttingen. My research focuses on multimodal AI, vision-to-time series modeling, and robust model evaluation. Skilled in deep learning, computer vision, diffusion models, and time-series analysis. Experienced with high-performance computing and large-scale, noisy datasets.

Hi there!👋 I’m Goirik Chakrabarty, a Ph.D. candidate in Data Science at the University of Göttingen. My research focuses on neuroscience data and mouse visual cortex. For the machine learning folks I want to develop a multimodal foundation model replicating the mouse visual cortex, this entails connecting vision and time-series data which is very learning and partially correlated. I work with deep learning, computer vision, diffusion models, transformers and time-series analysis, and I have practical experience with high-performance computing and large, complex datasets.

Prior to this, I worked as a Pre-Doctoral Researcher at TCS Research under Ramya Hebbalaguppe in collaboration with Prathosh A.P. (IISc Bangalore). Here I worked on vision-language models🤖, for image editing.

I completed my MS Thesis titled,”Continual Domain Incremental Learning during Test-time” under the supervision of Soma Biswas (IISc Bangalore) and Leelavati Narlikar (IISER Pune).

I am actively looking for Internship opportunities in AI/ML roles! Feel free to contact me 😄

selected publications

ICLR

OmniMouse: Scaling properties of multi-modal, multi-task Brain Models on 150B Neural Tokens

Konstantin Friedrich Willeke, Polina Turishcheva, Alex Gilbert, and 18 more authors

In International Conference on Learning Representations, 2026
NeurIPS-W

Beyond Pixels: A Differentiable Pipeline for Probing Neuronal Selectivity in 3D

Pavithra Elumalai, Mohammad Bashiri, Goirik Chakrabarty, and 2 more authors

In NeurIPS Workshop on Representational Alignment (NeurReps), 2025

Oral
ACMMM

LoMOE: Localized Multi-Object Editing via Multi-Diffusion

Goirik Chakrabarty, Aditya Chandrasekar, Ramya Hebbalaguppe, and 1 more author

In Proceedings of the 32nd ACM International Conference on Multimedia, 2024

Abs Code Website

Recent developments in diffusion models have demonstrated an exceptional capacity to generate high-quality, prompt-conditioned image edits. Nevertheless, previous approaches have primarily relied on textual prompts for image editing, which tend to be less effective when making precise edits to specific objects or fine-grained regions within a scene containing single/multiple objects. We introduce a novel framework for zero-shot localized multi-object editing through a multi-diffusion process to overcome this challenge. This framework empowers users to perform various operations on objects within an image, such as adding, replacing, or editing many objects in a complex scene in one pass. Our approach leverages foreground masks and corresponding simple text prompts that exert localized influences on the target regions resulting in high-fidelity image editing. A combination of cross-attention and background preservation losses within the latent space ensures that the characteristics of the object being edited are preserved while simultaneously achieving a high-quality, seamless reconstruction of the background with fewer artifacts compared to the state-of-the-art (SOTA). We also curate and release a dataset dedicated to multi-object editing, named LoMOE-Bench. Our experiments against existing SOTA demonstrate the improved effectiveness of our approach in terms of both image editing quality, and inference speed.
TMLR

SANTA: Source Anchoring Network and Target Alignment for Continual Test Time Adaptation

Goirik Chakrabarty, Manogna Sreenivas, and Soma Biswas

In Transactions on Machine Learning Research, 2023

Abs Code Website

Adapting a trained model to perform satisfactorily on continually changing test environments is an important and challenging task. In this work, we propose a novel framework, SANTA, which aims to satisfy the following characteristics required for online adaptation: 1) can work effectively for different (even small) batch sizes; 2) should continue to work well on the source domain; 3) should have minimal tunable hyperparameters and storage requirements. Given a pre-trained network trained on source domain data, the proposed framework modifies the affine parameters of the batch normalization layers using source anchoring based self-distillation. This ensures that the model incorporates knowledge from the newly encountered domains, without catastrophically forgetting the previously seen domains. We also propose a source-prototype driven contrastive alignment to ensure natural grouping of the target samples, while maintaining the already learnt semantic information. Extensive evaluation on three benchmark datasets under challenging settings justify the effectiveness of SANTA for real-world applications.