8 papers
Dr. Free: You Don't Need Difficulty Rewards for Self-Evolving Search Agents
Zhipeng Qian, Zihan Liang, Yufei Ma +10
A central limitation of current data-free self-evolution methods for training search agents is their reliance on difficulty-based proposer rewards. These methods reward a proposer…
OneRetrieval: Unifying Multi-Branch E-commerce Retrieval with an Editable Generative Model
Xuxin Zhang, Ben Chen, Yue Lv +13
Industrial e-commerce search serves hundreds of millions of items through a multi-branch retrieval stage fused by hand-tuned merging without joint optimization. Generative retrieva…
DV-SFT: Direct Vision Supervision for Fine-Grained Visual Understanding
Jianfei Zhao, Feng Zhang, Xin Sun +3
Multimodal large language models are typically trained end-to-end to predict ground-truth answers, yet supervision signals are applied exclusively to text tokens. Visual tokens, th…
SVFSearch: A Multimodal Knowledge-Intensive Benchmark for Short-Video Frame Search in the Gaming Vertical Domain
Lingtao Mao, Huangyu Dai, Xinyu Sun +4
Multimodal large language models are increasingly used as agent backbones that understand multimodal inputs, plan retrieval actions, invoke external tools, and reason over retrieve…
TIGER-FG: Text-Guided Implicit Fine-Grained Grounding for E-commerce Retrieval
Xinyu Sun, Huangyu Dai, Lingtao Mao +5
E-commerce image search often takes a cropped image as the query, while each candidate is represented by full item images and structured text. This image-to-multimodal retrieval se…
OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search Framework
Ben Chen, Siyuan Wang, Yufei Ma +20
Generative Retrieval (GR) has emerged as a promising paradigm for modern search systems. Compared to multi-stage cascaded architecture, it offers advantages such as end-to-end join…
UniDGF: A Unified Detection-to-Generation Framework for Hierarchical Object Visual Recognition
Xinyu Nan, Lingtao Mao, Huangyu Dai +8
Achieving visual semantic understanding requires a unified framework that simultaneously handles object detection, category prediction, and attribute recognition. However, current…
OneVision: An End-to-End Generative Framework for Multi-view E-commerce Vision Search
Zexin Zheng, Huangyu Dai, Lingtao Mao +8
Traditional vision search, similar to search and recommendation systems, follows the multi-stage cascading architecture (MCA) paradigm to balance efficiency and conversion. Specifi…