6 citations · 9 across the 11 of their papers we have counts for
13 papers
Language Model Fingerprinting Requires Rethinking Watermark Teachers
Jeongyeon Hwang, Anshul Nasery, Sewoong Oh +1
LLM fingerprinting via watermark distillation embeds a statistical watermark signal into model weights, enabling model owners to identify their models behind black-box APIs. Revisi…
MIRAGE: Protecting against Malicious Image Editing via False Moderation
Anshul Nasery, Ramnath Kumar, Cho-Jui Hsieh +1
The proliferation of AI-powered image editing systems raises serious concerns because it allows personal images to be arbitrarily manipulated at scale, with minimal effort, and a l…
Are Robust LLM Fingerprints Adversarially Robust?
Anshul Nasery, Edoardo Contente, Alkin Kaz +2
Model fingerprinting has emerged as a promising paradigm for claiming model ownership. However, robustness evaluations of these schemes have mostly focused on benign perturbations…
Scalable Fingerprinting of Large Language Models
Anshul Nasery, Jonathan Hayase, Creston Brooks +4
Model fingerprinting has emerged as a powerful tool for model owners to identify their shared model given API access. However, to lower false discovery rate, fight fingerprint leak…
OML: A Primitive for Reconciling Open Access with Owner Control in AI Model Distribution
Zerui Cheng, Edoardo Contente, Ben Finch +9
The current paradigm of AI model distribution presents a fundamental dichotomy: models are either closed and API-gated, sacrificing transparency and local execution, or openly dist…
PLeaS -- Merging Models with Permutations and Least Squares
Anshul Nasery, Jonathan Hayase, Pang Wei Koh +1
The democratization of machine learning systems has made the process of fine-tuning accessible to practitioners, leading to a wide range of open-source models fine-tuned on special…
PEEKABOO: Interactive Video Generation via Masked-Diffusion
Yash Jain, Anshul Nasery, Vibhav Vineet +1
Modern video generation models like Sora have achieved remarkable success in producing high-quality videos. However, a significant limitation is their inability to offer interactiv…
Label Differential Privacy via Aggregation
Anand Brahmbhatt, Rishi Saket, Shreyas Havaldar +3
This paper explores the use of linear aggregation to protect the privacy of sensitive training labels through the concept of \emph{label differential privacy} (label-DP) while main…
End-to-End Neural Network Compression via Regularized Latency Surrogates
Anshul Nasery, Hardik Shah, Arun Sai Suggala +1
Neural network (NN) compression via techniques such as pruning, quantization requires setting compression hyperparameters (e.g., number of channels to be pruned, bitwidths for quan…
Learning an Invertible Output Mapping Can Mitigate Simplicity Bias in Neural Networks
Sravanti Addepalli, Anshul Nasery, R. Venkatesh Babu +2
Deep Neural Networks are known to be brittle to even minor distribution shifts compared to the training distribution. While one line of work has demonstrated that Simplicity Bias (…
DAFT: Distilling Adversarially Fine-tuned Models for Better OOD Generalization
Anshul Nasery, Sravanti Addepalli, Praneeth Netrapalli +1
We consider the problem of OOD generalization, where the goal is to train a model that performs well on test distributions that are different from the training distribution. Deep l…
Training for the Future: A Simple Gradient Interpolation Loss to Generalize Along Time
Anshul Nasery, Soumyadeep Thakur, Vihari Piratla +2
In several real world applications, machine learning models are deployed to make predictions on data whose distribution changes gradually along time, leading to a drift between the…
Teaching CNNs to mimic Human Visual Cognitive Process & regularise Texture-Shape bias
Satyam Mohla, Anshul Nasery, Biplab Banerjee
Recent experiments in computer vision demonstrate texture bias as the primary reason for supreme results in models employing Convolutional Neural Networks (CNNs), conflicting with…