Researchers have developed CoGR, a novel retrieval framework that trains LLMs to generate retrieval representations for both queries and items. This approach uses a two-stage training process, starting with supervised fine-tuning to align keyword spaces, followed by co-evolving reinforcement learning. CoGR demonstrated superior performance on app marketplace and benchmark datasets, significantly outperforming existing methods. AI
IMPACT This framework could enhance search engine efficiency and relevance by enabling LLMs to directly generate retrieval representations.
RANK_REASON The cluster contains an academic paper detailing a new method for information retrieval.
Read on arXiv cs.IR (Information Retrieval) →
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
