{"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/spatial-r1-enhancing-mllms-in-video-spatial","title":"SpaceR: Reinforcing MLLMs in Video Spatial Reasoning","arxiv_id":"2504.01805","date":"2025-04-02","proceeding":null,"authors":["Kun Ouyang","Yuanxin Liu","HaoNing Wu","Yi Liu","Hao Zhou","Jie zhou","Fandong Meng","Xu sun"],"abstract":"Video spatial reasoning, which involves inferring the underlying spatial structure from observed video frames, poses a significant challenge for existing Multimodal Large Language Models (MLLMs). This limitation stems primarily from 1) the absence of high-quality datasets for this task, and 2) the lack of effective training strategies to develop spatial reasoning capabilities. Motivated by the success of Reinforcement Learning with Verifiable Reward (RLVR) in unlocking LLM reasoning abilities, this work aims to improve MLLMs in video spatial reasoning through the RLVR paradigm. To this end, we introduce the $\\textbf{SpaceR}$ framework. First, we present $\\textbf{SpaceR-151k}$, a dataset with 91k questions spanning diverse spatial reasoning scenarios with verifiable answers, and 60k samples for maintaining general multimodal understanding. Second, we propose $\\textbf{Spatially-Guided RLVR (SG-RLVR)}$, a novel reinforcement learning approach that extends Group Relative Policy Optimization (GRPO) with a novel map imagination mechanism, which encourages the model to infer spatial layouts in the thinking process, thereby facilitating more effective spatial reasoning. Extensive experiments demonstrate that SpaceR achieves state-of-the-art performance on spatial reasoning benchmarks (e.g., VSI-Bench, STI-Bench, and SPAR-Bench), while maintaining competitive results on video understanding benchmarks (e.g., Video-MME, TempCompass, and LongVideoBench). Remarkably, SpaceR surpasses the advanced GPT-4o by 11.6\\% accuracy on VSI-Bench and is on par with the leading proprietary model Gemini-2.0-Flash, highlighting the effectiveness of our SpaceR-151k dataset and SG-RLVR in reinforcing spatial reasoning ability of MLLMs. Code, model, and dataset are available at https://github.com/OuyangKun10/SpaceR.","url_abs":"https://arxiv.org/abs/2504.01805v2","url_pdf":"https://arxiv.org/pdf/2504.01805v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"spatial-r1-enhancing-mllms-in-video-spatial","repo_url":"https://github.com/ouyangkun10/spacer","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"spatial-r1-enhancing-mllms-in-video-spatial","repo_url":"https://github.com/ouyangkun10/spatial-r1","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"mme","task_name":"MME"},{"task_slug":"spatial-reasoning","task_name":"Spatial Reasoning"},{"task_slug":null,"task_name":"Video MME"},{"task_slug":"video-understanding","task_name":"Video Understanding"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2504.01805","atlas_url":"https://app.syntology.ai/?focus=2504.01805","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.01805"}},"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":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ouyangkun10/spacer","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ouyangkun10/spatial-r1","reach":{"status":"ok","spdx":"NOASSERTION"}}],"summary":{"ran_honours":1,"ran_draft_wrong":1},"by_repo_kind":{"official":{"samples":2,"ran":2,"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":2,"samples":[{"code_sha256_prefix":"f23cbd84b69ec52c","entry":"format_reward","repo":"ouyangkun10/spacer","repo_kind":"official","path":"SpaceR-SG-RLVR/src/r1-v/src/open_r1/SG-RLVR.py","file_url":"https://github.com/ouyangkun10/spacer/blob/HEAD/SpaceR-SG-RLVR/src/r1-v/src/open_r1/SG-RLVR.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"f23cbd84b69ec52c"}},{"code_sha256_prefix":"b7a3e714d1d33517","entry":"prepare_data","repo":"ouyangkun10/spacer","repo_kind":"official","path":"SpaceR-Eval/evaluate.py","file_url":"https://github.com/ouyangkun10/spacer/blob/HEAD/SpaceR-Eval/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"b7a3e714d1d33517"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}