AB
Alex Bie
cs.LGcs.CRstat.MLcs.CLcs.AIcs.DScs.CVcs.ITcs.SDeess.AS
On Valency
published · living versionsW_4amt7tv3·v1 · currentpublished
SimHack: Predicting the onset of detectable reward hacking before RL training
with Junwei Deng, Natalia Ponomareva
1 version
W_ev86av6e·v1published
Preprints & journals
18 papers in the corpus · 2019–2026SimHack: Predicting the onset of detectable reward hacking before RL training2610.28707v1 · Junwei Deng, Alex Bie, Natalia Ponomareva2026 · 0 citationsarXivon Valency
Actually, All Learnable Distribution Classes Are Privately Learnable2610.28878v1 · Mohammad Afzali, Hassan Ashtiani, Alex Bie et al.2026 · 0 citationsarXivon Valency
ContinuousBench: Can Differentially Private Synthetic Text Improve Capabilities?2606.01849v2 · Peihan Liu, Lucas Rosenblatt, Weiwei Kong et al.2026 · 0 citationsarXiv
AI-rithmetic2602.10416v1 · Alex Bie, Travis Dick, Alex Kulesza et al.2026 · 0 citationsarXiv
Learning from Synthetic Data: Limitations of ERM2601.15468v2 · Kareem Amin, Alex Bie, Weiwei Kong et al.2026 · 0 citationsarXiv
How to DP-fy Your Data: A Practical Guide to Generating Synthetic Data With Differential Privacy2512.03238v2 · Natalia Ponomareva, Zheng Xu, H. Brendan McMahan et al.2025 · 1 citationJAIR 2026, vol. 86 JAIR, Vol. 86
On the Learnability of Distribution Classes with Adaptive Adversaries2509.05137v1 · Tosca Lechner, Alex Bie, Gautam Kamath2025 · 0 citationsarXiv
Clustering and Median Aggregation Improve Differentially Private Inference2506.04566v1 · Kareem Amin, Salman Avestimehr, Sara Babakniya et al.2025 · 0 citationsarXiv
Escaping Collapse: The Strength of Weak Data for Large Language Model Training2502.08924v2 · Kareem Amin, Sara Babakniya, Alex Bie et al.2025 · 0 citationsarXiv
Private prediction for large-scale synthetic text generation2407.12108v2 · Kareem Amin, Alex Bie, Weiwei Kong et al.2024 · 5 citationsarXiv
Distribution Learnability and Robustness2406.17814v1 · Shai Ben-David, Alex Bie, Gautam Kamath et al.2024 · 1 citationarXiv
Parametric Feature Transfer: One-shot Federated Learning with Foundation Models2402.01862v1 · Mahdi Beitollahi, Alex Bie, Sobhan Hemati et al.2024 · 3 citationsarXiv
Understanding the Role of Layer Normalization in Label-Skewed Federated Learning2308.09565v2 · Guojun Zhang, Mahdi Beitollahi, Alex Bie et al.2023 · 0 citationsarXiv
Private Distribution Learning with Public Data: The View from Sample Compression2308.06239v2 · Shai Ben-David, Alex Bie, Cl'ement L. Canonne et al.2023 · 1 citationarXiv
Private GANs, Revisited2302.02936v2 · Alex Bie, Gautam Kamath, Guojun Zhang2023 · 4 citationsarXiv
Private Estimation with Public Data2208.07984v2 · Alex Bie, Gautam Kamath, Vikrant Singhal2022 · 8 citationsarXiv
Don't Generate Me: Training Differentially Private Generative Models with Sinkhorn Divergence2111.01177v2 · Tianshi Cao, Alex Bie, Arash Vahdat et al.2021 · 24 citationsAdvances in Neural Information Processing Systems, Volume 34, pages 12480--12492, year 2021
A Simplified Fully Quantized Transformer for End-to-end Speech Recognition1911.03604v4 · Alex Bie, Bharat Venkitesh, Joao Monteiro et al.2019 · 18 citationsarXiv
Career total: 25 works. 18 are in this corpus.
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.