AI Paper Atlas

Research map / Bandits and online learning / Online convex optimization and games

Next-Token Prediction and Regret Minimization

Mehryar Mohri, Clayton Sanford, Jon Schneider, Kiran Vodrahalli, Yifan Wu

ICML 2026 · Bandits and online learning › Online convex optimization and games · approach: Best-of-both-worlds bandit algorithms · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train on robustified (Polya-filtered, prefix-masked) data instead of plain BERNOULLINanoDO transformer · 8 ground-truth sequence distributionsregretImproved
robustify BERNOULLI modelNanoDO transformer · in-distribution Ber(1/3)->Ber(2/3)regret: about -0.16 retainedNo clear effect
use ROBUST_BERNOULLI instead of POLYAURN to stay near original distributionNanoDO transformer · TV-distance vs BERNOULLI_1TV-distance: 1.0000 -> 0.7602Improved
use ROBUST_BERNOULLI instead of POLYAURNNanoDO transformer · Next-token TV vs BERNOULLI_1Next-Token TV: 0.1529 -> 0.0199Improved
use ROBUST_BERNOULLI instead of POLYAURNNanoDO transformer · Next-token TV vs BERNOULLI_2Next-Token TV: 0.1655 -> 0.0299Improved
mask loss on high-regret prefix of Polya sequences (ablation)ROBUST_BERNOULLI training · training dataTV-distanceImproved
robustification vs seed-to-seed variationROBUST_BERNOULLI · full-sequence TVTV-distance: 0.4193 -> 0.7602Hurt
use a bounded context length equal to original modelrobustification (theory) · bounded contextregretHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy