The VCA framework uses diversity, consistency, and preference rewards to fine-tune a diffusion model with LoRA over multi-round dialogues, reporting improved intent alignment.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Optimizing Multi-Round Enhanced Training in Diffusion Models for Improved Preference Understanding
The VCA framework uses diversity, consistency, and preference rewards to fine-tune a diffusion model with LoRA over multi-round dialogues, reporting improved intent alignment.