conftrace_

data relabeling

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

llms (1) reward augmented (1) data resampling (1) training bia (1) quality score (1) cross-entropy loss (1) influence function (1) label noise (1) preference alignment (1) direct preference optimization (1)

Papers

Reward-Augmented Data Enhances Direct Preference Alignment of LLMs ICML 2025 Resolving Training Biases via Influence-based Data Relabeling ICLR 2022