SPARK: Self-Correcting Embodied Memory via Physical Re-Observation
Overview
This research presents SPARK, a corrective multi-episode framework for embodied retrieval-augmented generation (RAG) that addresses persistent errors in semantic labels, object positions, and spatial relationships.
Generated from attributed reports · Updated 2 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 · EQA questions: 158 other · reported trials
“viewpoint conditions, using perception-built KBs, a visibility-constrained oracle used only for evaluation, and a fixed set of 158 embodied question answering (EQA) questions scored before and after correction. A single corrective episode improves downstream EQA on every map under both conditions while storing 13-41% fewer objects than append-only updating, with gains up to 2.7x on one map-condition pair. Extending”
Exact source · revision 1Source owner not reported
Reported quantity · performance gain: 2.7 other · Basis not reported
“viewpoint conditions, using perception-built KBs, a visibility-constrained oracle used only for evaluation, and a fixed set of 158 embodied question answering (EQA) questions scored before and after correction. A single corrective episode improves downstream EQA on every map under both conditions while storing 13-41% fewer objects than append-only updating, with gains up to 2.7x on one map-condition pair. Extending”
Exact source · revision 1Source owner not reported
Reported quantity · maps evaluated: 4 other · reported trials
“urate object positions, and inconsistent spatial relationships. To address this, we present SPARK, a corrective multi-episode framework for embodied RAG. After the first episode, a semantic critic flags regions of the KB, a targeted planner generates revisits, and a selective merge adds, removes, or replaces KB information based on the new observations. We evaluate SPARK on four outdoor AirSim maps under two degraded”
Exact source · revision 1Source owner not reported
Reported quantity · episodes: 4 other · reported trials
“correction through Episode 4 brings every map above the EQA performance of its undegraded reference. Ablations show that the critic's three signals each contribute to performance, that targeted revisits achieve a larger mean gain than untargeted ones under the same budget, and that selective merging achieves gains comparable to append-only updating with a smaller KB. Overall, SPARK demonstrates the value of correctin”
Exact source · revision 1Source owner not reported
Reported quantity · critic signals: 3 other · Basis not reported
“correction through Episode 4 brings every map above the EQA performance of its undegraded reference. Ablations show that the critic's three signals each contribute to performance, that targeted revisits achieve a larger mean gain than untargeted ones under the same budget, and that selective merging achieves gains comparable to append-only updating with a smaller KB. Overall, SPARK demonstrates the value of correctin”
Exact source · revision 1Source owner not reported
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalSPARK: Self-Correcting Embodied Memory via Physical Re-Observation
This research presents SPARK, a corrective multi-episode framework for embodied retrieval-augmented generation (RAG) that addresses persistent errors in semantic labels, object positions, and spatial relationships.
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).