conftrace_

kahneman-tversky optimization

2 papers

Explore in graph

Also known as

KTO

Co-occurring keywords · papers mentioning both

group-wise calibration (1) semi-online training (1) factuality alignment (1) long text generation (1) large language model (1) long-form generation (1) hallucination mitigation (1) model alignment (1) preference optimization (1) text generation (1)

Papers

Beyond Binary Preferences: Semi-Online Label-Free GRACE-KTO with Group-Wise Adaptive Calibration for High-Quality Long-Text Generation EMNLP 2025 FactAlign: Long-form Factuality Alignment of Large Language Models EMNLP 2024