10 papers
Panprediction: Optimal Predictions for Any Downstream Task and Loss
Sivaraman Balakrishnan, Nika Haghtalab, Daniel Hsu +2
Supervised learning is classically formulated as training a model to minimize a fixed loss function over a fixed distribution, or task. However, an emerging paradigm instead views…
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…
From Style to Facts: Mapping the Boundaries of Knowledge Injection with Finetuning
Eric Zhao, Pranjal Awasthi, Nika Haghtalab
Finetuning provides a scalable and cost-effective means of customizing language models for specific tasks or response styles, with greater reliability than prompting or in-context…
Truthfulness of Decision-Theoretic Calibration Measures
Mingda Qiao, Eric Zhao
Calibration measures quantify how much a forecaster's predictions violates calibration, which requires that forecasts are unbiased conditioning on the forecasted probabilities. Two…
Sample, Scrutinize and Scale: Effective Inference-Time Search by Scaling Verification
Eric Zhao, Pranjal Awasthi, Sreenivas Gollapudi
Sampling-based search, a simple paradigm for utilizing test-time compute, involves generating multiple candidate responses and selecting the best one -- typically by having models…
Learning Variational Inequalities from Data: Fast Generalization Rates under Strong Monotonicity
Eric Zhao, Tatjana Chavdarova, Michael Jordan
Variational inequalities (VIs) are a broad class of optimization problems encompassing machine learning problems ranging from standard convex minimization to more complex scenarios…