[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86098-en":3,"doc-seo-86098-105":29,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":11,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},86098,2336464648746,"Skyler","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","TreeSoc: Tree-Structured Dynamic Reasoning and Tool Synergy for Soccer Video Understanding","Automated understanding of complex soccer scenarios from video is a major challenge for current vision-language models, which often show shallow cross-modal alignment, limited multi-step reasoning, and weak coordinated tool usage. TreeSoc reformulates soccer video question answering as a hierarchical search over a reasoning tree, using dynamic depth-first search to decompose queries into ordered subtasks with explicit intermediate states. The framework adaptively routes domain-specific modules and integrates their outputs at each node. On SoccerBench it reaches 85.2%/87.4%/82.2% on TextQA/ImageQA/VideoQA, and 74.16% on NExT-QA.","arXiv :2607 . 10990v1 [ cs .CV] 13 Jul 2026  \nTreeSoc: Tree-Structured Dynamic Reasoning and Tool Synergy for Soccer Video Understanding  \nThanh-Nhan Vo 1,2 , Thanh-Khoi Nguyen 1,2 , Trong-Thuan Nguyen 1,2 , Trung-Hoang Le 1,2 , and Minh-Triet Tran 1,2  \n1 University of Science, VNU-HCM, Ho Chi Minh City, Vietnam  \n2Vietnam National University, Ho Chi Minh City, Vietnam  \nABSTRACT  \nAutomated understanding of complex soccer scenarios from video remains a significant challenge for contemporary vision-language models (VLMs), which suffer from shallow cross-modal alignment and exhibit fundamental limitations in multi-step reasoning and coordinated tool integration. We present TreeSoc, a structured reasoning framework that reformulates soccer video question answering as a hierarchical search problem rather than a single-pass prediction. Specifically, TreeSoc employs a dynamic depth-first search (DFS) mechanism that decomposes complex queries into sequentially ordered sub-tasks, enabling iterative reasoning refinement through explicit intermediate states. This tree-structured decomposition naturally supports adaptive tool routing, wherein domain-specific modules are selectively activated and their outputs incorporated at each reasoning node to produce contextually grounded predictions. On SoccerBench, TreeSoc achieves state-of-the-art performance, with accuracies of 85 .2%, 87 .4%, and 82 .2% on TextQA, ImageQA, and VideoQA, respectively. Additionally, TreeSoc further demonstrates strong cross-domain generalization, attaining 74 . 16% accuracy on NExT-QA. These results establish structured, tool-augmented tree reasoning as an effective paradigm for robust video understanding. Code is available at: [https://github.com/thanhnhan29/TreeSoc](https://github.com/thanhnhan29/TreeSoc).  \nKeywords: Video Question Answering, Soccer Video Understanding, Tree-Search Reasoning, Multimodal Large Language Models (MLLMs), Multi-Agent Systems  \n1. INTRODUCTION  \nVisual Question Answering (VQA) has emerged as an important paradigm for extracting semantic and contextual information from visual data, particularly in dynamic video environments [1] . In the sports domain, especially soccer, VQA supports a wide range of specialized analytical tasks, including foul recognition, player tracking, tactical interpretation, and commentary generation [2, 3 , 4] . Beyond answering factual questions, high-level soccer VQA requires models to reason over complex spatio-temporal events, such as tactical buildups, player interactions, and the causal factors underlying refereeing decisions [3] . These characteristics make soccer video understanding a challenging yet valuable foundation for studying domain-specific visual reasoning.  \nDespite significant progress in video understanding and sport analysis, existing vision-language models (VLMs) still face notable limitations when applied to complex sports VQA tasks. In particular, most current models rely on single-pass inference and implicit visual-textual alignment, which limits their capacity to construct explicit reasoning trajectories, coordinate specialized perception tools, and revise intermediate decisions [5, 6] . These limitations are particularly critical in soccer videos, where relevant evidence is often distributed across multiple video frames, grounded in domain-specific concepts, and dependent on external knowledge sources or specialized analytical modules [7, 8] . Therefore, early visual misinterpretations may propagate directly to the final answer when the reasoning process lacks structured verification and adaptive replanning mechanisms [9] .  \nIn this paper, we propose TreeSoc, a structured reasoning framework for visual question answering over soccer videos. Rather than treating VQA as a single-pass prediction problem, TreeSoc reformulates it as a multi-step state-space search over a hierarchical reasoning tree [10] . The framework employs a Multimodal Large Language Model (MLLM) as a central coordin","cbCaibUPwbDBkNL5","https://ap.wps.com/l/cbCaibUPwbDBkNL5","pdf",1758555,4,1,"English","en",105,"# Introduction\n# Related Work","[{\"question\":\"What core approach does TreeSoc use for soccer video question answering?\",\"answer\":\"TreeSoc reformulates soccer VQA as a hierarchical, tree-structured state-space search rather than a single-pass prediction, enabling multi-step reasoning over ordered subtasks.\"},{\"question\":\"How does TreeSoc perform reasoning refinement across multiple steps?\",\"answer\":\"TreeSoc uses a dynamic depth-first search that decomposes complex queries, updates execution based on intermediate observations, and revises its reasoning path to reduce error propagation.\"},{\"question\":\"Which tools or external information sources does TreeSoc integrate?\",\"answer\":\"TreeSoc adaptively invokes domain-expert modules (e.g., YOLO26, PRTREID, UniSoccer) and can retrieve relevant information from external databases, incorporating their outputs into reasoning nodes.\"}]",1784208491,20,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"treesoc-tree-structured-dynamic-reasoning-and-tool-synergy-for-soccer-video-understanding","",{"@graph":35,"@context":84},[36,52,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":20},"https://docshare.wps.com/document/treesoc-tree-structured-dynamic-reasoning-and-tool-synergy-for-soccer-video-understanding/86098/",{"url":51,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":23,"description":14,"dateModified":60,"datePublished":61,"encodingFormat":59,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":40,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-07-26","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":20},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What core approach does TreeSoc use for soccer video question answering?","Question",{"text":74,"@type":75},"TreeSoc reformulates soccer VQA as a hierarchical, tree-structured state-space search rather than a single-pass prediction, enabling multi-step reasoning over ordered subtasks.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does TreeSoc perform reasoning refinement across multiple steps?",{"text":79,"@type":75},"TreeSoc uses a dynamic depth-first search that decomposes complex queries, updates execution based on intermediate observations, and revises its reasoning path to reduce error propagation.",{"name":81,"@type":72,"acceptedAnswer":82},"Which tools or external information sources does TreeSoc integrate?",{"text":83,"@type":75},"TreeSoc adaptively invokes domain-expert modules (e.g., YOLO26, PRTREID, UniSoccer) and can retrieve relevant information from external databases, incorporating their outputs into reasoning nodes.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,126,129,133],{"id":21,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":28,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":28,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":28,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]