{"paper":{"title":"Think3D: Thinking with Space for Spatial Reasoning","arxiv_id":"2601.13029","primary_cat":"cs.CV","submitted_at":"2026-01-19T13:13:54+00:00","version_compiled":"v4"},"notice":"Every quoted sentence is the authors' text from the arXiv listing abstract. Labels and probabilities are the outputs of classifier-1.13.0 on the question bank abstract-p1-20260923, exactly as returned. No sentence in this record was written by a language model.","schema":"pith.compiled-record/v0","status":"compiled","abstract":{"source":"arXiv listing abstract v4","sentences":[{"i":0,"p":0.57,"id":"s0","own":false,"role":"background","text":"While Vision-Language Models (VLMs) excel at 2D visual understanding, they remain constrained by 2D-centric paradigm that severely limits genuine 3D spatial reasoning.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.09,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.41,"role_probabilities":{"gap":0.43,"method":0.0,"result":0.0,"background":0.57,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":1,"p":1.0,"id":"s1","own":true,"role":"contribution","text":"To bridge this gap, we introduce Think3D, a novel framework that equips VLM agents with interactive, 3D chain-of-thought reasoning capabilities.","hedge":{"p":0.98,"label":"asserted"},"own_p":0.88,"main_claim_p":0.05,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":1.0,"interpretation":0.0}},{"i":2,"p":0.86,"id":"s2","own":true,"role":"method","text":"By integrating a suite of 3D manipulation tools, Think3D transforms perception into active spatial exploration, mirroring human geometric reasoning.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.62,"main_claim_p":0.07,"quantitative":false,"quantitative_p":0.32,"role_probabilities":{"gap":0.0,"method":0.86,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.14,"interpretation":0.0}},{"i":3,"p":1.0,"id":"s3","own":true,"role":"result","text":"Think3D consistently improves proprietary models, including GPT-4.1 and Gemini 2.5 Pro, across BLINK Multi-view, MindCube-1K, and VSI-Bench-Tiny.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.91,"main_claim_p":0.09,"quantitative":false,"quantitative_p":0.49,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":4,"p":0.91,"id":"s4","own":true,"role":"contribution","text":"We further propose Think3D-RL to teach smaller open-weight models how to manipulate 3D space effectively.","hedge":{"p":0.96,"label":"asserted"},"own_p":0.88,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.07,"role_probabilities":{"gap":0.0,"method":0.09,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.91,"interpretation":0.0}},{"i":5,"p":0.87,"id":"s5","own":true,"role":"method","text":"Using only final-answer rewards, without process supervision or handcrafted exploration trajectories, Think3D-RL enables Qwen3-VL-4B to autonomously learn effective 3D exploration strategies.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.05,"quantitative":true,"quantitative_p":0.61,"role_probabilities":{"gap":0.0,"method":0.87,"result":0.07,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.06,"interpretation":0.0}},{"i":6,"p":1.0,"id":"s6","own":true,"role":"result","text":"After training, the model exhibits tool-use patterns similar to those of stronger proprietary models, while shifting the effect of 3D tool use on MindCube-1K from a performance drop to a substantial improvement.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.1,"quantitative":true,"quantitative_p":0.69,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":7,"p":1.0,"id":"s7","own":true,"role":"interpretation","text":"These results show that active exploration in 3D space provides an effective and general paradigm for improving spatial reasoning in multimodal agents.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.85,"main_claim_p":0.9,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":1.0}},{"i":8,"p":0.68,"id":"s8","own":false,"role":"contribution","text":"Code, models, and data are available at https://github.com/zhangzaibin/spagent.","hedge":{"p":0.87,"label":"asserted"},"own_p":0.33,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.03,"result":0.01,"background":0.02,"limitation":0.0,"future_work":0.25,"contribution":0.68,"interpretation":0.0}}]},"compiled":{"main_claim":{"p":1.0,"text":"These results show that active exploration in 3D space provides an effective and general paradigm for improving spatial reasoning in multimodal agents.","location":"abstract s7","hedge_strength":{"p":1.0,"label":"asserted"}},"provenance":{"route":{"via":"TypeSafe API","model":"classifier-1.13.0","endpoint":"https://api.typesafe.ai/v1/systemone","provider":"TypeSafe"},"packing":"one request per paper","requests":1,"input_tokens":6993,"cost_usd_reported":0.0},"compiled_at":"2026-09-23","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.14,"states_limitation":0.16,"human_participants":0.04,"controlled_comparison":0.49,"randomized_assignment":0.02,"states_quantitative_result":0.12,"compares_with_prior_results":0.77,"mentions_code_or_data_availability":0.99},"evidential_basis":{"p":0.98,"value":"observation or experiment","probabilities":{"argument":0.0,"literature":0.0,"proof_or_derivation":0.0,"analysis_of_existing_data":0.0,"observation_or_experiment":0.98,"simulation_or_computation":0.02}},"contribution_type":{"p":1.0,"value":"method or tool","probabilities":{"review":0.0,"method_or_tool":1.0,"empirical_study":0.0,"simulation_study":0.0,"theoretical_result":0.0,"dataset_or_benchmark":0.0,"position_or_commentary":0.0}},"contribution_statement":{"p":0.99,"text":"To bridge this gap, we introduce Think3D, a novel framework that equips VLM agents with interactive, 3D chain-of-thought reasoning capabilities.","location":"abstract s1"}},"coverage":{"tier":"abstract","levels_run":["P abstract"],"not_examined":["full text","figures and tables","references and citing sentences","relations to other papers"],"sources_read":["arXiv listing abstract v4"],"levels_not_run":["S sections","T body sentences","C citations","R cross-paper relations"],"sentences_examined":9,"sentences_in_abstract":9},"measured_accuracy":{"url":"https://pith.science/accuracy","measured_on":"2026-10-02","method":"two independent model families labelled a fresh random sample blind; the figure is how often the published label matches their consensus","sentence_role_agreement":0.92,"main_claim_agreement":0.908,"typed_citation_agreement":0.933,"typed_citation_agreement_95_interval":[0.917,0.945],"review_comments_accurate":0.758,"review_comments_accurate_95_interval":[0.684,0.82],"review_caught_admitted_error_found_or_related":0.448,"review_comments_judged":{"judged_by":["gpt-5.6-luna","grok-4.7"],"judged_against":"Pith's own text copy of each paper: text extracted from the PDF, the first 90,000 characters","measured_on":"2026-10-04","accurate":113,"decided":149,"sampled_reviews":200,"split_by_lane":false,"by_lane":"https://pith.science/accuracy#by-lane"}}}