{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/arxiv-2511-00405","title":"UME-R1: Exploring Reasoning-Driven Generative Multimodal Embeddings","arxiv_id":"2511.00405","date":"2025-11-01","proceeding":"ICLR","authors":["Zhibin Lan","Liqiang Niu","Fandong Meng","Jie Zhou","Jinsong Su"],"abstract":"The remarkable success of multimodal large language models (MLLMs) has driven advances in multimodal embeddings, yet existing models remain inherently discriminative, limiting their ability to benefit from reasoning-driven generation paradigm. In this work, we pioneer the exploration of generative embeddings, unifying embedding tasks within a generative paradigm. We propose UME-R1, a universal multimodal embedding framework consisting of a two-stage training strategy: a cold-start supervised fine-tuning equips the model with reasoning capabilities and enables it to generate both discriminative and generative embeddings; a subsequent reinforcement learning enhances reasoning and further optimizes generative embedding quality. This pioneering work reveals four key insights: 1) generative embeddings unlock substantial performance gains over conventional discriminative embeddings by leveraging the powerful generative reasoning capabilities of MLLMs; 2) discriminative and generative embeddings are complementary, whose combined oracle performance far exceeding that of either alone; 3) RL can effectively enhance generative embeddings, establishing a scalable optimization paradigm.; 4) repeated sampling at inference boosts downstream task coverage (pass@k), highlighting the inference-time scalability potential of generative embeddings. Evaluated on the MMEB-V2 benchmark across 78 tasks spanning video, image, and visual documents, UME-R1 significantly outperforms conventional discriminative embedding models and offers a foundation for more interpretable, reasoning-driven generative multimodal embeddings. Our code, models, and datasets will be publicly available at https://github.com/XMUDeepLIT/UME-R1.","url_abs":"https://arxiv.org/abs/2511.00405","url_pdf":"https://arxiv.org/pdf/2511.00405","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2511.00405","atlas_url":"https://app.syntology.ai/?focus=2511.00405","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2511.00405"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/XMUDeepLIT/UME-R1","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran_fixture":1,"unverified":14},"by_repo_kind":{"found_in_text":{"samples":15,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"3f31b48fd862935d","entry":"get_image_data_url","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/r1-train/local_scripts/create_vision_cot_data.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/r1-train/local_scripts/create_vision_cot_data.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3f31b48fd862935d"}},{"code_sha256_prefix":"41c548a25e90eb39","entry":"extract_problem_solution","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/r1-train/local_scripts/prepare_hf_data.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/r1-train/local_scripts/prepare_hf_data.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"41c548a25e90eb39"}},{"code_sha256_prefix":"981eb19d513b4f7e","entry":"extract_step","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/eval/VLM2Vec/adhoc/gather_score_byckpt_aws.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/eval/VLM2Vec/adhoc/gather_score_byckpt_aws.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"981eb19d513b4f7e"}},{"code_sha256_prefix":"349fd11ea2ede40c","entry":"get_disc_embedding_idx","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/eval/VLM2Vec/eval_twomode.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/eval/VLM2Vec/eval_twomode.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"349fd11ea2ede40c"}},{"code_sha256_prefix":"f2c74df1482c3bcc","entry":"get_gen_embedding_idx","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/eval/VLM2Vec/eval_twomode.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/eval/VLM2Vec/eval_twomode.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f2c74df1482c3bcc"}},{"code_sha256_prefix":"58ab41ce28a94c62","entry":"gpt4o_query","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/r1-train/local_scripts/create_vision_cot_data.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/r1-train/local_scripts/create_vision_cot_data.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"58ab41ce28a94c62"}},{"code_sha256_prefix":"2bf49beebed5b7dc","entry":"load_frames","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/data-download/llavahound_cp_download.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/data-download/llavahound_cp_download.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2bf49beebed5b7dc"}},{"code_sha256_prefix":"bdcc94560fb87f30","entry":"load_image_from_path","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/r1-train/local_scripts/prepare_hf_data.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/r1-train/local_scripts/prepare_hf_data.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"bdcc94560fb87f30"}},{"code_sha256_prefix":"4ea7e0c6070c077d","entry":"load_vidore_dataset","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/data-download/vidore_download.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/data-download/vidore_download.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4ea7e0c6070c077d"}},{"code_sha256_prefix":"afc674648320a7d1","entry":"load_visrag_dataset","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/data-download/visrag_download.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/data-download/visrag_download.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"afc674648320a7d1"}},{"code_sha256_prefix":"cbeb923f8c89fca4","entry":"normalize_reps","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"demo/image_disc_embed_demo.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/demo/image_disc_embed_demo.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"cbeb923f8c89fca4"}},{"code_sha256_prefix":"8fd49856d19d27fc","entry":"process_query","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/data-download/visrag_download.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/data-download/visrag_download.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8fd49856d19d27fc"}},{"code_sha256_prefix":"f029d6b5f19d2885","entry":"process_single_item","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/r1-train/local_scripts/create_vision_cot_data.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/r1-train/local_scripts/create_vision_cot_data.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f029d6b5f19d2885"}},{"code_sha256_prefix":"0d43c38f16bd6dd8","entry":"process_video_frames","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/data-download/llavahound_cp_download.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/data-download/llavahound_cp_download.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0d43c38f16bd6dd8"}},{"code_sha256_prefix":"726f29532082b33f","entry":"sample_frames","repo":"XMUDeepLIT/UME-R1","repo_kind":"found_in_text","path":"src/data-download/llavahound_cp_download.py","file_url":"https://github.com/XMUDeepLIT/UME-R1/blob/HEAD/src/data-download/llavahound_cp_download.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"726f29532082b33f"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.LG","source":"arxiv_api"},"syntology_extracted_results":null}