Large Reasoning Embedding Models: Towards Next-Generation Dense Retrieval Paradigm
Preprint
Abstract
In modern e-commerce search systems, dense retrieval has become an indispensable component. By computing similarities between query and item (product) embeddings, it efficiently selects candidate products from large-scale repositories. With the breakthroughs in large language models (LLMs), mainstream embedding models have gradually shifted from BERT to LLMs for more accurate text modeling. However, these models still adopt direct-embedding methods, and the semantic accuracy of embeddings remains inadequate. Therefore, contrastive learning is heavily employed to achieve tight semantic alignment between positive pairs. Consequently, such models tend to capture statistical co-occurrence patterns in the training data, biasing them toward shallow lexical and semantic matches. For difficult queries exhibiting notable lexical disparity from target items, the performance degrades significantly. In this work, we propose the Large Reasoning Embedding Model (LREM), which novelly integrates reasoning processes into representation learning. For difficult queries, LREM first conducts reasoning to achieve a deep understanding of the original query, and then produces a reasoning-augmented query embedding for retrieval. This reasoning process effectively bridges the semantic gap between original queries and target items, significantly improving retrieval accuracy. Specifically, we adopt a two-stage training process: the first stage optimizes the LLM on carefully curated Query-CoT-Item triplets with SFT and InfoNCE losses to establish preliminary reasoning and embedding capabilities, and the second stage further refines the reasoning trajectories via reinforcement learning (RL). Extensive offline and online experiments validate the effectiveness of LREM, leading to its deployment on China's largest e-commerce platform since August 2025.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Can We Further Elicit Reasoning in LLMs? Critic-Guided Planning with Retrieval-Augmentation for Solving Challenging Tasks2025
- OneGen: Efficient One-Pass Unified Generation and Retrieval for LLMs2024
- AutoCAP: Towards Automatic Cross-lingual Alignment Planning for Zero-shot Chain-of-Thought2024
- MathPrompter: Mathematical Reasoning using Large Language Models2023
- RetroMAE: Pre-Training Retrieval-oriented Language Models Via Masked Auto-Encoder2022
- A Deep Look into neural ranking models for information retrievalInformation Processing & Management, 2020
- BERT Rediscovers the Classical NLP Pipeline2019
- Evaluating collaborative filtering recommender systemsACM Transactions on Information Systems, 2004
- Measuring Search Engine QualityInformation Retrieval Journal, 2001
- An optimal algorithm for approximate nearest neighbor searching fixed dimensionsJournal of the ACM, 1998