FEIJOA: Task-Driven Grounding and Execution for Mobile Manipulation
Overview
The paper introduces FEIJOA, a language-model agent that grounds instructions at test time using RGB-D scans and verifies skill execution. On 103 open-ended grounding queries, it resolves 94.2% exactly versus 68.0% for the best prebuilt representation. Over 50 real-world trials on Spot, it completes 35 tasks; 17 successes follow recovery from a failed tool call or verification check. Of 15 failed trials, 13 were caused by physical skills and only 2 by grounding or reasoning.
Generated from attributed reports · Updated 1 hours ago
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · real-world trials on Spot: 50 other · reported trials
“nded grounding queries, FEIJOA resolves 94.2% exactly, compared with 68.0% for the best prebuilt representation. Over 50 real-world trials on Spot, it completes 35 tasks; 17 of these successes follow recovery from a failed tool call or verification check. Of the 15 failed trials, 13 were caused by the robot's physical skills, and only 2 by grounding or the agent's reasoning. Project website: https://williamshen-nz.gi”
Exact source · revision 1Source owner not reported
Reported quantity · tasks completed: 35 other · reported trials
“nded grounding queries, FEIJOA resolves 94.2% exactly, compared with 68.0% for the best prebuilt representation. Over 50 real-world trials on Spot, it completes 35 tasks; 17 of these successes follow recovery from a failed tool call or verification check. Of the 15 failed trials, 13 were caused by the robot's physical skills, and only 2 by grounding or the agent's reasoning. Project website: https://williamshen-nz.gi”
Exact source · revision 1Source owner not reported
Reported quantity · successes following recovery from failure: 17 other · reported trials
“nded grounding queries, FEIJOA resolves 94.2% exactly, compared with 68.0% for the best prebuilt representation. Over 50 real-world trials on Spot, it completes 35 tasks; 17 of these successes follow recovery from a failed tool call or verification check. Of the 15 failed trials, 13 were caused by the robot's physical skills, and only 2 by grounding or the agent's reasoning. Project website: https://williamshen-nz.gi”
Exact source · revision 1Source owner not reported
Reported quantity · failed trials caused by robot physical skills: 13 other · reported trials
“nded grounding queries, FEIJOA resolves 94.2% exactly, compared with 68.0% for the best prebuilt representation. Over 50 real-world trials on Spot, it completes 35 tasks; 17 of these successes follow recovery from a failed tool call or verification check. Of the 15 failed trials, 13 were caused by the robot's physical skills, and only 2 by grounding or the agent's reasoning. Project website: https://williamshen-nz.gi”
Exact source · revision 1Source owner not reported
Reported quantity · failed trials caused by grounding or agent reasoning: 2 other · reported trials
“nded grounding queries, FEIJOA resolves 94.2% exactly, compared with 68.0% for the best prebuilt representation. Over 50 real-world trials on Spot, it completes 35 tasks; 17 of these successes follow recovery from a failed tool call or verification check. Of the 15 failed trials, 13 were caused by the robot's physical skills, and only 2 by grounding or the agent's reasoning. Project website: https://williamshen-nz.gi”
Exact source · revision 1Source owner not reported
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalFEIJOA: Task-Driven Grounding and Execution for Mobile Manipulation
The paper introduces FEIJOA, a language-model agent that grounds instructions at test time using RGB-D scans and verifies skill execution. On 103 open-ended grounding queries, it resolves 94.2% exactly versus 68.0% for the best prebuilt representation. Over 50 real-world trials on Spot, it completes 35 tasks; 17 successes follow recovery from a failed tool call or verification check. Of 15 failed trials, 13 were caused by physical skills and only 2 by grounding or reasoning.
Event coverage history
There is not enough continuous observation data to show a trend.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).