About

Hi, I am Samyadeep Basu, a 3rd year CS PhD student at UMD, College Park (2022 - Present). I work with Soheil Feizi in the Center for Machine Learning. My research focus is reliable deep learning encompassing methods to learn from data with limited supervision (few-shot learning), model interpretability (how to understand decision making process of deep networks), robustness and more recently vision+language. Previously I received my MS from UMD in 2020 and then spent close to two years at Microsoft AI in the ML rotation program. During my stint at Microsoft AI, I worked with the Language Science Team at Azure AI and MSAI where I researched, developed and deployed large-scale language models for various scenarios. My latest CV can be accessed at CV-Link.

News

(January 2024) : Paper on diffusion model interpretability accepted to ICLR 2024!

(December 2023) : Paper on few-shot finetuning for vision models accepted to AAAI 2024!

(August-October 2023): Paper on surgical fine-tuning for “small” language models accepted to EMNLP 2023!

(July 2023): Excited to announce two new projects : (i) Improving CLIP using knowledge distillation from diffusion models; (ii) Benchmark for text-guided image editing methods!

(June 2023): Started internship at Adobe Research! Working on interpretability + model editing for text-to-image generative models!

(April 2023): Our new pre-print on PEFT modules for few-shot fine-tuning is on arXiv!

(Jan 2023): Paper on algorithm to design difficult few-shot tasks accepted at ICLR 2023!

(September 2022): Finished an amazing research internship at Microsoft Research working with Daniela Massiceti on few-shot learning!

(Feb 2022): Started my PhD to work on few-shot learning and model interpretability!

(October 2020 - Jan 2022) : Break from Grad School to work at Microsoft AI as an Applied Scientist!

(Feb 2021): Papers on influence functions at ICLR 2021 and ICML 2020!

Publications

  1. Localizing and Editing Knowledge in Text-to-Image Models

    ICLR 2024

    We propose an interpretability framework + model editing method to ablate concepts from text-to-image models, fast!

  2. Augmenting CLIP with Improved Visio-Linguistic Reasoning

    Under Review

    We propose a knowledge-distillation technique to improve reasoning abilities in CLIP!

  3. EditVal: Benchmarking Diffusion Based Text-Guided Image Editing Methods

    Under Review Code

    We propose a new comprehensive benchmark for evaluating diffusion based editing methods!

  4. On Surgical Finetuning for Language Encoders

    EMNLP 2023

    Method to surgically finetune language encoders with a subset of layers to perform close to full-finetuning!

  5. Strong Baselines for Parameter-Efficient Few-Shot Fine-Tuning

    AAAI 2024 Code

    We propose two easy-to-implement strong baselines for PEFT which leads to SoTA on MD!

  6. Hard Meta-Dataset++: Towards Understanding Few-shot Performance on Difficult Tasks

    ICLR 2023 Code / Media Coverage

    We propose a fast algorithm - FastDiffSel which can extract difficult few-shot tasks in a computational efficient way from large vision datasets!

  7. Strategies to Improve Few-Shot Learning for Intent Classification and Slot Filling

    NAACL 2022 (SUKI)

    Propose empirical strategies to improve few-shot performance for joint intent classification and slot-filling.

  8. Influence Functions in Deep Learning are Fragile

    ICLR 2021

    End to end analysis of influence functions in deep learning!

  9. On Second-Order Group Influence Functions for Black-Box Predictions

    ICML 2020

    We propose second-order group influence functions, which are better suited to handle group effects!

  10. Membership Model Inversion Attacks for Deep Networks

    NeurIPS 2020(w)

    We propose an early inversion technique using GANs to do membership inference!