AI Paper Atlas

Research map / Reinforcement learning / RL from feedback

Learning Correlated Reward Models: Statistical Barriers and Opportunities

Yeshwanth Cherapanamjeri, Constantinos Costis Daskalakis, Gabriele Farina, Sobhan Mohammadpour

ICLR 2026 · Reinforcement learning › RL from feedback · approach: Theory of RLHF · 2 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 the probit with best-of-three instead of pairwise observationscorrelated probit RUM · Synthetic (mu=0, Sigma=bin)median accuracy: 0.50 -> 0.79Improved
train the probit with best-of-three instead of pairwise observationscorrelated probit RUM · Synthetic (mu=0, Sigma=r)median accuracy: 0.51 -> 0.67Improved
train the probit with best-of-three instead of pairwise observationscorrelated probit RUM · Synthetic (mu=r, Sigma=I)median accuracy: 0.66 -> 0.70Improved
train the probit with best-of-three instead of pairwise observationscorrelated probit RUM · Synthetic (mu=r, Sigma=bin)median accuracy: 0.95 -> 0.95No clear effect
train the probit with best-of-three instead of pairwise observationscorrelated probit RUM · Sushi-Amedian accuracy: 0.65 -> 0.68Improved
train the probit with best-of-three instead of pairwise observationscorrelated probit RUM · Netflix 150k (onehot)median accuracy: 0.56 -> 0.61Improved
train the probit with best-of-three instead of pairwise observationscorrelated probit RUM · MovieLens 50k (LLM features)median accuracy: 0.55 -> 0.59Improved
use correlated probit (best-of-three) instead of logitreward/choice model · Sushi-Amedian accuracy: 0.66 -> 0.68Improved
use correlated probit (best-of-three) instead of logitreward/choice model · Netflix 150k (onehot)median accuracy: 0.61 -> 0.61No clear effect
use correlated probit (best-of-three) instead of logitreward/choice model · Synthetic (mu=0, Sigma=bin)median accuracy: 0.50 -> 0.79Improved
use LLM embedding features instead of onehot featuresprobit (best-of-three) · Netflix 150kmedian accuracy: 0.61 -> 0.61No clear effect
use LLM embedding features instead of onehot featuresprobit (best-of-three) · MovieLens 50kmedian accuracy: 0.59 -> 0.59No clear effect

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