collaborators

6 papers

cs.LG2026

Optimal Deterministic Multicalibration and Omniprediction

Georgy Noarov, Aaron Roth

A model is multicalibrated on a collection of group weights if it is calibrated -- i.e. unbiased even conditional on its prediction -- not just overall, but also after reweight…

cs.LG2026

Optimal Lower Bounds for Online Multicalibration

Natalie Collina, Jiuyao Lu, Georgy Noarov +1

We prove tight lower bounds for online multicalibration, establishing an information-theoretic separation from marginal calibration. In the general setting where group functions ca…

cs.LG2026

The Sample Complexity of Multicalibration

Natalie Collina, Jiuyao Lu, Georgy Noarov +1

We study the minimax sample complexity of multicalibration in the batch setting. A learner observes i.i.d. samples from an unknown distribution and must output a (possibly rand…

cs.AI2026

Foundations of Top- Decoding For Language Models

Georgy Noarov, Soham Mallick, Tao Wang +5

Top- decoding is a widely used method for sampling from LLMs: at each token, only the largest next-token-probabilities are kept, and the next token is sampled after re-norma…

cs.AI2026

Statistical Early Stopping for Reasoning Models

Yangxinyu Xie, Tao Wang, Soham Mallick +6

While LLMs have seen substantial improvement in reasoning capabilities, they also sometimes overthink, generating unnecessary reasoning steps, particularly under uncertainty, given…

stat.ME2025

Stronger Neyman Regret Guarantees for Adaptive Experimental Design

Georgy Noarov, Riccardo Fogliato, Martin Bertran +1

We study the design of adaptive, sequential experiments for unbiased average treatment effect (ATE) estimation in the design-based potential outcomes setting. Our goal is to develo…