no code implementations • 19 Mar 2024 • Qilong Wu, Varun Chandrasekaran
SCTS obtains color information by strategically prompting the watermarked LLM and comparing output tokens frequencies.
no code implementations • 6 Mar 2024 • Rishabh Adiga, Lakshminarayanan Subramanian, Varun Chandrasekaran
This approach avoids the training of a dedicated model for selection of examples, and instead uses certain metrics to align the syntactico-semantic complexity of test sentences and examples.
no code implementations • 25 Oct 2023 • Fan Wu, Huseyin A. Inan, Arturs Backurs, Varun Chandrasekaran, Janardhan Kulkarni, Robert Sim
Positioned between pre-training and user deployment, aligning large language models (LLMs) through reinforcement learning (RL) has emerged as a prevailing strategy for training instruction following-models such as ChatGPT.
1 code implementation • 24 Oct 2023 • Marah I Abdin, Suriya Gunasekar, Varun Chandrasekaran, Jerry Li, Mert Yuksekgonul, Rahee Ghosh Peshawaria, Ranjita Naik, Besmira Nushi
Motivated by rising concerns around factual incorrectness and hallucinations of LLMs, we present KITAB, a new dataset for measuring constraint satisfaction abilities of language models.
no code implementations • 12 Oct 2023 • Jihye Choi, Shruti Tople, Varun Chandrasekaran, Somesh Jha
Many practical black-box MIAs require query access to the data distribution (the same distribution where the private data is drawn) to train shadow models.
no code implementations • 11 Oct 2023 • Ranjita Naik, Varun Chandrasekaran, Mert Yuksekgonul, Hamid Palangi, Besmira Nushi
Large language models (LLMs) are documented to struggle in settings that require complex reasoning.
no code implementations • 10 Oct 2023 • Erik Jones, Hamid Palangi, Clarisse Simões, Varun Chandrasekaran, Subhabrata Mukherjee, Arindam Mitra, Ahmed Awadallah, Ece Kamar
We also find that optimizing the system message rather than the model weights can be critical; fine-tuning the entire model on the synthetic task can counterintuitively increase hallucination.
1 code implementation • 26 Sep 2023 • Mert Yuksekgonul, Varun Chandrasekaran, Erik Jones, Suriya Gunasekar, Ranjita Naik, Hamid Palangi, Ece Kamar, Besmira Nushi
We investigate the internal behavior of Transformer-based Large Language Models (LLMs) when they generate factually incorrect text.
2 code implementations • 22 Mar 2023 • Sébastien Bubeck, Varun Chandrasekaran, Ronen Eldan, Johannes Gehrke, Eric Horvitz, Ece Kamar, Peter Lee, Yin Tat Lee, Yuanzhi Li, Scott Lundberg, Harsha Nori, Hamid Palangi, Marco Tulio Ribeiro, Yi Zhang
We contend that (this early version of) GPT-4 is part of a new cohort of LLMs (along with ChatGPT and Google's PaLM for example) that exhibit more general intelligence than previous AI models.
Ranked #33 on Arithmetic Reasoning on GSM8K
1 code implementation • 17 Oct 2022 • Thorsten Eisenhofer, Doreen Riepel, Varun Chandrasekaran, Esha Ghosh, Olga Ohrimenko, Nicolas Papernot
In this framework, the server first computes a proof that the model was trained on a dataset $D$.
no code implementations • 6 Aug 2022 • Congyu Fang, Hengrui Jia, Anvith Thudi, Mohammad Yaghini, Christopher A. Choquette-Choo, Natalie Dullerud, Varun Chandrasekaran, Nicolas Papernot
They empirically argued the benefit of this approach by showing how spoofing--computing a proof for a stolen model--is as expensive as obtaining the proof honestly by training the model.
no code implementations • 10 Jun 2022 • Varun Chandrasekaran, Suman Banerjee, Diego Perino, Nicolas Kourtellis
Federated learning (FL), where data remains at the federated clients, and where only gradient updates are shared with a central aggregator, was assumed to be private.
1 code implementation • 27 Sep 2021 • Anvith Thudi, Gabriel Deza, Varun Chandrasekaran, Nicolas Papernot
In this work, we first taxonomize approaches and metrics of approximate unlearning.
no code implementations • 20 Sep 2021 • Varun Chandrasekaran, Hengrui Jia, Anvith Thudi, Adelin Travers, Mohammad Yaghini, Nicolas Papernot
The application of machine learning (ML) in computer systems introduces not only many benefits but also risks to society.
no code implementations • 3 Aug 2021 • Adelin Travers, Lorna Licollari, Guanghan Wang, Varun Chandrasekaran, Adam Dziedzic, David Lie, Nicolas Papernot
In the white-box setting, we instantiate this class with a joint, multi-stage optimization attack.
no code implementations • 27 May 2021 • Varun Chandrasekaran, Darren Edge, Somesh Jha, Amit Sharma, Cheng Zhang, Shruti Tople
However for real-world applications, the privacy of data is critical.
2 code implementations • 9 Mar 2021 • Hengrui Jia, Mohammad Yaghini, Christopher A. Choquette-Choo, Natalie Dullerud, Anvith Thudi, Varun Chandrasekaran, Nicolas Papernot
In particular, our analyses and experiments show that an adversary seeking to illegitimately manufacture a proof-of-learning needs to perform *at least* as much work than is needed for gradient descent itself.
1 code implementation • 29 Jul 2020 • Jayaram Raghuram, Varun Chandrasekaran, Somesh Jha, Suman Banerjee
We propose an unsupervised anomaly detection framework based on the internal DNN layer representations in the form of a meta-algorithm with configurable components.
1 code implementation • 19 Mar 2020 • Chuhan Gao, Varun Chandrasekaran, Kassem Fawaz, Somesh Jha
We implement and evaluate Face-Off to find that it deceives three commercial face recognition services from Microsoft, Amazon, and Face++.
Cryptography and Security
3 code implementations • 27 Feb 2020 • Hengrui Jia, Christopher A. Choquette-Choo, Varun Chandrasekaran, Nicolas Papernot
Such pairs are watermarks, which are not sampled from the task distribution and are only known to the defender.
1 code implementation • 26 Feb 2020 • Sanghyun Hong, Varun Chandrasekaran, Yiğitcan Kaya, Tudor Dumitraş, Nicolas Papernot
In this work, we study the feasibility of an attack-agnostic defense relying on artifacts that are common to all poisoning attacks.
2 code implementations • 9 Dec 2019 • Lucas Bourtoule, Varun Chandrasekaran, Christopher A. Choquette-Choo, Hengrui Jia, Adelin Travers, Baiwu Zhang, David Lie, Nicolas Papernot
Once users have shared their data online, it is generally difficult for them to revoke access and ask for the data to be deleted.
no code implementations • 2 Oct 2019 • Lakshya Jain, Wilson Wu, Steven Chen, Uyeong Jang, Varun Chandrasekaran, Sanjit Seshia, Somesh Jha
In this paper we explore semantic adversarial examples (SAEs) where an attacker creates perturbations in the semantic space representing the environment that produces input for the ML model.
no code implementations • 26 May 2019 • Varun Chandrasekaran, Brian Tang, Nicolas Papernot, Kassem Fawaz, Somesh Jha, Xi Wu
and how to design a classification paradigm that leverages these invariances to improve the robustness accuracy trade-off?
no code implementations • 5 Nov 2018 • Varun Chandrasekaran, Kamalika Chaudhuri, Irene Giacomelli, Somesh Jha, Songbai Yan
This has resulted in the surge of Machine Learning-as-a-Service (MLaaS) - cloud services that provide (a) tools and resources to learn the model, and (b) a user-friendly query interface to access the model.