3 papers
cs.LG2025
The Limits of Preference Data for Post-Training
Eric Zhao, Jessica Dai, Pranjal Awasthi
Recent progress in strengthening the capabilities of large language models has stemmed from applying reinforcement learning to domains with automatically verifiable outcomes. A key…
cs.LG2024
Learning With Multi-Group Guarantees For Clusterable Subpopulations
Jessica Dai, Nika Haghtalab, Eric Zhao
A canonical desideratum for prediction problems is that performance guarantees should hold not just on average over the population, but also for meaningful subpopulations within th…
cs.LG2024
Can Probabilistic Feedback Drive User Impacts in Online Platforms?
Jessica Dai, Bailey Flanigan, Nika Haghtalab +2
A common explanation for negative user impacts of content recommender systems is misalignment between the platform's objective and user welfare. In this work, we show that misalign…