SOURCE-LINKED INTELLIGENCE

Learning from What You Retrieve: Online RL Fine-Tuning for Semantic Retrieval

arXiv · AI, language, vision and robotics · article · Aug 31, 2026 · UTC

In large-scale e-commerce retrieval, dual-encoder retrievers are op- timized for contrastive similarity, whereas downstream rerankers capture finer-grained relevance preferences; this objective mis- match limits end-to-end retrieval quality. Reinforcement Learning offers a way to use reward-model feedback for retriever adaptation, but we observe that standard policy-gradient updates can degrade embedding geometry, especially when the document index must remain frozen due to industrial constraints. To address this, we propose PAO (Positive-Advantage-Only), a selective RL optimization method. Ou

Read original source ↗ Open in workspace

recordType

paper

region

Global