{"ID":2863510,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.24621","arxiv_id":"2509.24621","title":"FreeRet: MLLMs as Training-Free Retrievers","abstract":"Multimodal large language models (MLLMs) are emerging as versatile foundations for mixed-modality retrieval. Yet, they often require heavy post-hoc training to convert them into contrastive encoders for retrieval. This work asks: Can off-the-shelf MLLMs serve as powerful retrievers without additional training? We present FreeRet, a plug-and-play framework that turns any MLLM into a two-stage retriever. FreeRet first derives semantically grounded embeddings directly from the model for fast candidate search, and then exploits its reasoning ability for precise reranking. The framework contributes three advances: bypassing lexical alignment layers to obtain semantically faithful embeddings, conditioning representation generation with explicit priors, and mitigating framing effect in reranking via neutral choice framing. On the MMEB and MMEB-V2 benchmarks spanning 46 datasets, FreeRet substantially outperforms models trained on millions of pairs. Beyond benchmarks, FreeRet is model-agnostic and scales seamlessly across MLLM families and sizes, preserves their generative abilities, supports arbitrary modality combinations, and unifies retrieval, reranking, and generation into end-to-end RAG within a single model. Our findings demonstrate that pretrained MLLMs, when carefully harnessed, can serve as strong retrieval engines without training, closing a critical gap in their role as generalists.","short_abstract":"Multimodal large language models (MLLMs) are emerging as versatile foundations for mixed-modality retrieval. Yet, they often require heavy post-hoc training to convert them into contrastive encoders for retrieval. This work asks: Can off-the-shelf MLLMs serve as powerful retrievers without additional training? We prese...","url_abs":"https://arxiv.org/abs/2509.24621","url_pdf":"https://arxiv.org/pdf/2509.24621v3","authors":"[\"Yuhan Zhu\",\"Xiangyu Zeng\",\"Chenting Wang\",\"Xinhao Li\",\"Chunxu Liu\",\"Yicheng Xu\",\"Ziang Yan\",\"Yi Wang\",\"Limin Wang\"]","published":"2025-09-29T11:28:42Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}