BB
Boaz Barak
cs.CCcs.LGcs.DSstat.MLquant-phcs.AIcs.CLcs.CRcs.NE68W40 (Analysis of algorithms)
On Valency
published · living versionsW_jsb5733n·v1 · currentpublished
Deliberative Alignment: Reasoning Enables Safer Language Models
with Melody Y. Guan, Manas Joglekar, Eric Wallace, Saachi Jain +10
1 version
Preprints & journals
37 papers in the corpus · 2008–2025OpenAI o1 System Card2412.16720v2 · OpenAI: Aaron Jaech, Adam Kalai, Adam Lerer et al.2024 · 44 citationsarXiv
Stress Testing Deliberative Alignment for Anti-Scheming Training2509.15541v1 · Bronson Schoen, Evgenia Nitishinskaya, Mikita Balesni et al.2025 · 2 citationsarXiv
Watermarks in the Sand: Impossibility of Strong Watermarking for Generative Models2311.04378v5 · Hanlin Zhang, Benjamin L. Edelman, Danilo Francati et al.2023 · 9 citationsarXiv
Deliberative Alignment: Reasoning Enables Safer Language Models2412.16339v2 · Melody Y. Guan, Manas Joglekar, Eric Wallace et al.2024 · 34 citationsarXivon Valency
An Economic Solution to Copyright Challenges of Generative AI2404.13964v4 · Jiachen T. Wang, Zhun Deng, Hiroaki Chiba-Okabe et al.2024 · 6 citationsarXiv
Beyond Implicit Bias: The Insignificance of SGD Noise in Online Learning2306.08590v2 · Nikhil Vyas, Depen Morwani, Rosie Zhao et al.2023 · 1 citationarXiv
Limitations of Linear Cross-Entropy as a Measure for Quantum Advantage2112.01657v1 · Xun Gao, Marcin Kalinowski, Chi-Ning Chou et al.2021 · 70 citationsarXiv
Distinguishing the Knowable from the Unknowable with Language Models2402.03563v2 · Gustaf Ahdritz, Tian Qin, Nikhil Vyas et al.2024 · 3 citationsarXiv
Remote inspection of adversary-controlled environments.37848432 · Tobisch, Johannes, Philippe, Sébastien, Barak, Boaz et al.2023 · 3 citationsNature communications. 2023;14(1):6566
On Provable Copyright Protection for Generative Models2302.10870v2 · Nikhil Vyas, Sham Kakade, Boaz Barak2023 · 21 citationsarXiv
On the works of Avi Wigderson2307.09524v1 · Boaz Barak, Yael Kalai, Ran Raz et al.2023 · 0 citationsarXiv
Robust reconstruction of single-cell RNA-seq data with iterative gene weight updates.37387155 · Sheng, Yueqi, Barak, Boaz, Nitzan, Mor2023 · 1 citationBioinformatics (Oxford, England). 2023;39(39 Suppl 1):i423-i430
Named Tensor Notation2102.13196v3 · David Chiang, Alexander M. Rush, Boaz Barak2021 · 2 citationsTMLR, January 2023
Noisy tensor completion via the sum-of-squares hierarchy.35702694 · Barak, Boaz, Moitra, Ankur2022 · 46 citationsMathematical programming. 2022;193(2):513-548
Quantum Optimization of Maximum Independent Set using Rydberg Atom Arrays2202.09372v1 · Sepehr Ebadi, Alexander Keesling, Madelyn Cain et al.2022 · 471 citationsScience 376, 1209 (2022)
Deconstructing Distributions: A Pointwise Framework of Learning2202.09931v2 · Gal Kaplun, Nikhil Ghosh, Saurabh Garg et al.2022 · 1 citationarXiv
Classical algorithms and quantum limitations for maximum cut on high-girth graphs2106.05900v1 · Boaz Barak, Kunal Marwaha2021 · 19 citations13th Innovations in Theoretical Computer Science Conference (ITCS 2022); Article No. 14
Playing Unique Games on Certified Small-Set Expanders2006.09969v3 · Mitali Bafna, Boaz Barak, Pravesh Kothari et al.2020 · 23 citationsarXiv
For self-supervised learning, Rationality implies generalization, provably2010.08508v1 · Yamini Bansal, Gal Kaplun, Boaz Barak2020 · 12 citationsarXiv
Spoofing Linear Cross-Entropy Benchmarking in Shallow Quantum Circuits2005.02421v1 · Boaz Barak, Chi-Ning Chou, Xun Gao2020 · 19 citationsarXiv
On Higher-Order Cryptography (Long Version)2002.07218v1 · Boaz Barak, Raphaelle Crubill'e, Ugo Dal Lago2020 · 0 citationsarXiv
SGD on Neural Networks Learns Functions of Increasing Complexity1905.11604v1 · Preetum Nakkiran, Gal Kaplun, Dimitris Kalimeris et al.2019 · 103 citationsarXiv
Merkle's Key Agreement Protocol is Optimal: An $O(n^2)$ Attack on any Key Agreement from Random Oracles0801.3669v4 · Boaz Barak, Mohammad Mahmoody2008 · 0 citationsarXiv
Lower Bounds on Signatures from Symmetric Primitives0801.3680v3 · Boaz Barak, Mohammad Mahmoody2008 · 55 citationsarXiv
Small-Set Expansion in Shortcode Graph and the 2-to-2 Conjecture1804.08662v1 · Boaz Barak, Pravesh K. Kothari, David Steurer2018 · 32 citationsarXiv
Quantum entanglement, sum of squares, and the log rank conjecture1701.06321v2 · Boaz Barak, Pravesh Kothari, David Steurer2017 · 24 citationsarXiv
Making the long code shorter, with applications to the Unique Games Conjecture1111.0405v1 · Boaz Barak, Parikshit Gopalan, Johan Hastad et al.2011 · 20 citationsarXiv
A Nearly Tight Sum-of-Squares Lower Bound for the Planted Clique Problem1604.03084v2 · Boaz Barak, Samuel B. Hopkins, Jonathan Kelner et al.2016 · 173 citationsarXiv
Noisy Tensor Completion via the Sum-of-Squares Hierarchy1501.06521v3 · Boaz Barak, Ankur Moitra2015 · 46 citationsarXiv
Beating the random assignment on constraint satisfaction problems of bounded degree1505.03424v2 · Boaz Barak, Ankur Moitra, Ryan O'Donnell et al.2015 · 53 citationsarXiv
Dictionary Learning and Tensor Decomposition via the Sum-of-Squares Method1407.1543v2 · Boaz Barak, Jonathan A. Kelner, David Steurer2014 · 136 citationsarXiv
Sum-of-squares proofs and the quest toward optimal algorithms1404.5236v2 · Boaz Barak, David Steurer2014 · 108 citationsarXiv
Rounding Sum-of-Squares Relaxations1312.6652v1 · Boaz Barak, Jonathan Kelner, David Steurer2013 · 104 citationsarXiv
Rounding Semidefinite Programming Hierarchies via Global Correlation1104.4680v1 · Boaz Barak, Prasad Raghavendra, David Steurer2011 · 128 citationsarXiv
Rank Bounds for Design Matrices with Applications to Combinatorial Geometry and Locally Correctable Codes1009.4375v2 · Boaz Barak, Zeev Dvir, Avi Wigderson et al.2010 · 43 citationsarXiv
Subsampling Mathematical Relaxations and Average-case Complexity0911.5526v2 · Boaz Barak, Moritz Hardt, Thomas Holenstein et al.2009 · 26 citationsarXiv
Career total: 190 works. 37 are in this corpus.Showing the 36 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.