Selecting preference pairs whose DPO implicit reward gap is small yields better LLM alignment than random or baseline selection while using only 10% of the data.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2025 2verdicts
UNVERDICTED 2representative citing papers
JAM aligns frozen vision and language models via joint autoencoders and multimodal Spread Loss, reliably inducing cross-modal alignment across layer depths, objectives, and model scales.
citing papers explorer
-
Difficulty-Based Preference Data Selection by DPO Implicit Reward Gap
Selecting preference pairs whose DPO implicit reward gap is small yields better LLM alignment than random or baseline selection while using only 10% of the data.
-
Escaping Plato's Cave: JAM for Aligning Independently Trained Vision and Language Models
JAM aligns frozen vision and language models via joint autoencoders and multimodal Spread Loss, reliably inducing cross-modal alignment across layer depths, objectives, and model scales.