no code implementations • NAACL (SUKI) 2022 • Samyadeep Basu, Amr Sharaf, Karine Ip Kiun Chong, Alex Fischer, Vishal Rohra, Michael Amoake, Hazem El-Hammamy, Ehi Nosakhare, Vijay Ramani, Benjamin Han
Intent classification (IC) and slot filling (SF) are two fundamental tasks in modern Natural Language Understanding (NLU) systems.
no code implementations • 2 May 2024 • Samyadeep Basu, Keivan Rezaei, Priyatham Kattakinda, Ryan Rossi, Cherry Zhao, Vlad Morariu, Varun Manjunatha, Soheil Feizi
To address this issue, we introduce the concept of Mechanistic Localization in text-to-image models, where knowledge about various visual attributes (e. g., "style", "objects", "facts") can be mechanistically localized to a small fraction of layers in the UNet, thus facilitating efficient model editing.
no code implementations • 11 Apr 2024 • Mazda Moayeri, Samyadeep Basu, Sriram Balasubramanian, Priyatham Kattakinda, Atoosa Chengini, Robert Brauneis, Soheil Feizi
Recent text-to-image generative models such as Stable Diffusion are extremely adept at mimicking and generating copyrighted content, raising concerns amongst artists that their unique styles may be improperly copied.
no code implementations • 26 Oct 2023 • Reshmi Ghosh, Harjeet Singh Kajal, Sharanya Kamath, Dhuri Shrivastava, Samyadeep Basu, Hansi Zeng, Soundararajan Srinivasan
However, current works on topic segmentation often focus on segmentation of structured texts.
1 code implementation • 25 Oct 2023 • Abhilasha Lodha, Gayatri Belapurkar, Saloni Chalkapurkar, Yuanming Tao, Reshmi Ghosh, Samyadeep Basu, Dmitrii Petrov, Soundararajan Srinivasan
We show evidence that for different downstream language tasks, fine-tuning only a subset of layers is sufficient to obtain performance that is close to and often better than fine-tuning all the layers in the language encoder.
no code implementations • 20 Oct 2023 • Samyadeep Basu, Nanxuan Zhao, Vlad Morariu, Soheil Feizi, Varun Manjunatha
We adapt Causal Mediation Analysis for text-to-image models and trace knowledge about distinct visual attributes to various (causal) components in the (i) UNet and (ii) text-encoder of the diffusion model.
no code implementations • 3 Oct 2023 • Samyadeep Basu, Mehrdad Saberi, Shweta Bhardwaj, Atoosa Malemir Chegini, Daniela Massiceti, Maziar Sanjabi, Shell Xu Hu, Soheil Feizi
From both the human study and automated evaluation, we find that: (i) Instruct-Pix2Pix, Null-Text and SINE are the top-performing methods averaged across different edit types, however {\it only} Instruct-Pix2Pix and Null-Text are able to preserve original image properties; (ii) Most of the editing methods fail at edits involving spatial operations (e. g., changing the position of an object).
no code implementations • 18 Jul 2023 • Samyadeep Basu, Maziar Sanjabi, Daniela Massiceti, Shell Xu Hu, Soheil Feizi
On the challenging Winoground compositional reasoning benchmark, our method improves the absolute visio-linguistic performance of different CLIP models by up to 7%, while on the ARO dataset, our method improves the visio-linguistic performance by upto 3%.
no code implementations • 4 Apr 2023 • Samyadeep Basu, Daniela Massiceti, Shell Xu Hu, Soheil Feizi
Through our controlled empirical study, we have two main findings: (i) Fine-tuning just the LayerNorm parameters (which we call LN-Tune) during few-shot adaptation is an extremely strong baseline across ViTs pre-trained with both self-supervised and supervised objectives, (ii) For self-supervised ViTs, we find that simply learning a set of scaling parameters for each attention matrix (which we call AttnScale) along with a domain-residual adapter (DRA) module leads to state-of-the-art performance (while being $\sim\!$ 9$\times$ more parameter-efficient) on MD.
no code implementations • 27 Nov 2022 • Reshmi Ghosh, Harjeet Singh Kajal, Sharanya Kamath, Dhuri Shrivastava, Samyadeep Basu, Soundararajan Srinivasan
Breaking down a document or a conversation into multiple contiguous segments based on its semantic structure is an important and challenging problem in NLP, which can assist many downstream tasks.
no code implementations • 21 Oct 2021 • Samyadeep Basu, Amr Sharaf, Nicolo Fusi, Soheil Feizi
To address the issue of sub-par performance on hard episodes, we investigate and benchmark different meta-training strategies based on adversarial training and curriculum learning.
no code implementations • 17 Sep 2021 • Samyadeep Basu, Karine lp Kiun Chong, Amr Sharaf, Alex Fischer, Vishal Rohra, Michael Amoake, Hazem El-Hammamy, Ehi Nosakhare, Vijay Ramani, Benjamin Han
Intent classification (IC) and slot filling (SF) are two fundamental tasks in modern Natural Language Understanding (NLU) systems.
no code implementations • ICLR 2021 • Samyadeep Basu, Philip Pope, Soheil Feizi
Influence functions approximate the effect of training samples in test-time predictions and have a wide variety of applications in machine learning interpretability and uncertainty estimation.
no code implementations • 18 Nov 2019 • Rauf Izmailov, Peter Lin, Chris Mesterharm, Samyadeep Basu
We consider membership inference attacks, one of the main privacy issues in machine learning.
no code implementations • ICML 2020 • Samyadeep Basu, Xuchen You, Soheil Feizi
Often we want to identify an influential group of training samples in a particular test prediction for a given machine learning model.
no code implementations • 9 Oct 2019 • Samyadeep Basu, Rauf Izmailov, Chris Mesterharm
With the increasing adoption of AI, inherent security and privacy vulnerabilities formachine learning systems are being discovered.