DynamicPO prevents preference optimization collapse in multi-negative DPO by adaptively selecting boundary-critical negatives and calibrating per-sample optimization strength, yielding higher recommendation accuracy on three public datasets.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
LLMs show below-average consistency and vulnerability to false beliefs in emotional queries with false presuppositions, more so for moderate emotions.
citing papers explorer
-
DynamicPO: Dynamic Preference Optimization for Recommendation
DynamicPO prevents preference optimization collapse in multi-negative DPO by adaptively selecting boundary-critical negatives and calibrating per-sample optimization strength, yielding higher recommendation accuracy on three public datasets.
-
Towards Emotion Consistency Analysis of Large Language Models in Emotional Conversational Contexts
LLMs show below-average consistency and vulnerability to false beliefs in emotional queries with false presuppositions, more so for moderate emotions.