{"paper":{"title":"When Robots Do the Chores: A Benchmark and Agent for Long-Horizon Household Task Execution","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"HoloMind agent with DAG planner and dual memories raises long-horizon household task success while cutting dependence on model size.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bowen Pang, Jing Liu, Longteng Guo, Ruyi Ji, Xingjian He, Yanghong Mei, Zilin Zhu, Zongxun Zhang","submitted_at":"2026-05-14T07:47:53Z","abstract_excerpt":"Long-horizon household tasks demand robust high-level planning and sustained reasoning capabilities, which are largely overlooked by existing embodied AI benchmarks that emphasize short-horizon navigation or manipulation and rely on fixed task categories. We introduce LongAct, a benchmark designed to evaluate planning-level autonomy in long-horizon household tasks specified through free-form instructions. By abstracting away embodiment-specific low-level control, LongAct isolates high-level cognitive capabilities such as instruction understanding, dependency management, memory maintenance, and"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"HoloMind substantially improves long-horizon performance while reducing reliance on model scale. Even top models achieve only 59% goal completion and 16% full-task success.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"Abstracting away embodiment-specific low-level control isolates high-level cognitive capabilities such as instruction understanding, dependency management, memory maintenance, and adaptive planning without losing essential task realism.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"LongAct benchmark reveals top VLMs reach only 59% goal completion and 16% full success on long-horizon household tasks, while HoloMind agent improves results via DAG planner, multimodal spatial memory, episodic memory, and global critic.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"HoloMind agent with DAG planner and dual memories raises long-horizon household task success while cutting dependence on model size.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"ec12b714591e2539c6a20c56a8798200702d62b6d8ea32579e2375c6d12dd29d"},"source":{"id":"2605.14504","kind":"arxiv","version":1},"verdict":{"id":"94bc80ce-e9cf-49aa-b028-9585781dec0f","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T01:55:07.807981Z","strongest_claim":"HoloMind substantially improves long-horizon performance while reducing reliance on model scale. Even top models achieve only 59% goal completion and 16% full-task success.","one_line_summary":"LongAct benchmark reveals top VLMs reach only 59% goal completion and 16% full success on long-horizon household tasks, while HoloMind agent improves results via DAG planner, multimodal spatial memory, episodic memory, and global critic.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"Abstracting away embodiment-specific low-level control isolates high-level cognitive capabilities such as instruction understanding, dependency management, memory maintenance, and adaptive planning without losing essential task realism.","pith_extraction_headline":"HoloMind agent with DAG planner and dual memories raises long-horizon household task success while cutting dependence on model size."},"references":{"count":43,"sample":[{"doi":"","year":2025,"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","ref_index":1,"cited_arxiv_id":"2505.09388","is_internal_anchor":true},{"doi":"","year":2023,"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","ref_index":2,"cited_arxiv_id":"2303.08774","is_internal_anchor":true},{"doi":"","year":2023,"title":"H. Liu, C. Li, Q. Wu, and Y . J. Lee. Visual instruction tuning.Advances in Neural Information Processing Systems, 36:1–19, 2023","work_id":"4eec8a8e-8c6e-46ee-8008-9b1dd58415a1","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond","work_id":"cbc2bb21-b6bb-46c0-80bf-107e195ffe10","ref_index":4,"cited_arxiv_id":"2308.12966","is_internal_anchor":true},{"doi":"","year":2023,"title":"Gemini: A Family of Highly Capable Multimodal Models","work_id":"83f7c85b-3f11-450f-ac0c-64d9745220b2","ref_index":5,"cited_arxiv_id":"2312.11805","is_internal_anchor":true}],"resolved_work":43,"snapshot_sha256":"21571ef359ea4e79e652bc800ce523debebc34db345e2391734a67fe6616f61d","internal_anchors":12},"formal_canon":{"evidence_count":2,"snapshot_sha256":"d547d575035ac5376155c4db5e45601102cffa2ea632f260cf140bd9dc162946"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}