conftrace_

on-policy sampling

2 papers

Explore in graph

Co-occurring keywords

direct preference optimization (317) language model alignment (142) reward model (251) data efficiency (124) monte carlo estimator (15) preference datum (28) iterative training (38) alignment training (15) off-policy sampling (2) policy evaluation (115)

Papers

Finding the Sweet Spot: Preference Data Construction for Scaling Preference Optimization ACL 2025

Robust On-Policy Sampling for Data-Efficient Policy Evaluation in Reinforcement Learning NIPS 2022