TaskGround: Structured Executable Task Inference for Full-Scene Household Reasoning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Feng, ZhiYuan, Deng, Yu, An, Ruichuan, Liu, Zhenhua, Li, Qixiu, Wu, Keming, Du, Zhiying, Wang, Weijie, Wang, Haoxiao, Chen, Shuang, Xu, Sicheng, Liang, Yaobo, Yang, Jiaolong, Guo, Baining
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917507835625472
author Feng, ZhiYuan
Deng, Yu
An, Ruichuan
Liu, Zhenhua
Li, Qixiu
Wu, Keming
Du, Zhiying
Wang, Weijie
Wang, Haoxiao
Chen, Shuang
Xu, Sicheng
Liang, Yaobo
Yang, Jiaolong
Guo, Baining
author_facet Feng, ZhiYuan
Deng, Yu
An, Ruichuan
Liu, Zhenhua
Li, Qixiu
Wu, Keming
Du, Zhiying
Wang, Weijie
Wang, Haoxiao
Chen, Shuang
Xu, Sicheng
Liang, Yaobo
Yang, Jiaolong
Guo, Baining
contents In real home deployments, household agents must often operate from a complete household scene and a situated household request, rather than from a clean task specification. Such requests require agents to identify task-relevant entities, recover intended task conditions, and resolve ordering constraints from the surrounding scene context. We formalize this capability as full-scene household reasoning: given a complete household scene and a situated household request, an agent must infer executable task structure before producing a grounded skill-level action sequence. This setting is challenging because complete household scenes contain substantial task-irrelevant information, making direct complete-scene prompting inefficient and error-prone. In practical deployment, this challenge is further amplified by privacy and local compute constraints, which favor compact open-weight models with limited long-context reasoning ability. We propose TaskGround, a training-free and model-agnostic Ground-Infer-Execute framework that grounds complete scenes into compact task-relevant scene slices, infers executable task structure, and compiles it into grounded skill-level action sequences. To evaluate this setting, we introduce FullHome, a human-validated evaluation suite of 400 household tasks spanning diverse home-scale environments and both goal-oriented and process-constrained requirements. On FullHome, TaskGround improves task success rates by large margins across both proprietary and open-weight models. Notably, it makes Qwen3.5-9B competitive with GPT-5 under direct complete-scene prompting while reducing total input-token cost by up to 18x. Our results identify executable task-structure inference as a central bottleneck in full-scene household reasoning and show that structured grounding can make compact local models substantially more effective for practical household deployment.
format Preprint
id arxiv_https___arxiv_org_abs_2605_18109
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle TaskGround: Structured Executable Task Inference for Full-Scene Household Reasoning
Feng, ZhiYuan
Deng, Yu
An, Ruichuan
Liu, Zhenhua
Li, Qixiu
Wu, Keming
Du, Zhiying
Wang, Weijie
Wang, Haoxiao
Chen, Shuang
Xu, Sicheng
Liang, Yaobo
Yang, Jiaolong
Guo, Baining
Artificial Intelligence
Computer Vision and Pattern Recognition
Robotics
In real home deployments, household agents must often operate from a complete household scene and a situated household request, rather than from a clean task specification. Such requests require agents to identify task-relevant entities, recover intended task conditions, and resolve ordering constraints from the surrounding scene context. We formalize this capability as full-scene household reasoning: given a complete household scene and a situated household request, an agent must infer executable task structure before producing a grounded skill-level action sequence. This setting is challenging because complete household scenes contain substantial task-irrelevant information, making direct complete-scene prompting inefficient and error-prone. In practical deployment, this challenge is further amplified by privacy and local compute constraints, which favor compact open-weight models with limited long-context reasoning ability. We propose TaskGround, a training-free and model-agnostic Ground-Infer-Execute framework that grounds complete scenes into compact task-relevant scene slices, infers executable task structure, and compiles it into grounded skill-level action sequences. To evaluate this setting, we introduce FullHome, a human-validated evaluation suite of 400 household tasks spanning diverse home-scale environments and both goal-oriented and process-constrained requirements. On FullHome, TaskGround improves task success rates by large margins across both proprietary and open-weight models. Notably, it makes Qwen3.5-9B competitive with GPT-5 under direct complete-scene prompting while reducing total input-token cost by up to 18x. Our results identify executable task-structure inference as a central bottleneck in full-scene household reasoning and show that structured grounding can make compact local models substantially more effective for practical household deployment.
title TaskGround: Structured Executable Task Inference for Full-Scene Household Reasoning
topic Artificial Intelligence
Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2605.18109