RESEARCH · RESEARCH · #1407
GAVA: Grounded arbitration of user corrections in text-based embodied agents (arXiv:2610.00282v1)
The paper formulates 'grounded correction arbitration' — a decision framework for agents to accept, reject, inspect, or ask about potentially incorrect user corrections — and introduces GAVA, which uses observation-bounded evidence, legal probes, and a one-step expected-loss rule. Evaluated in a text-only ALFWorld benchmark across many checkpoints and scenarios, GAVA achieves near-perfect correction accuracy while lowering interaction cost via selective inspection informed by an object-location prior, but the study is limited to symbolic observations, controlled speakers, and no human, visual, or robotic evaluation.
KEY POINTS
- The paper formulates 'grounded correction arbitration' — a decision framework for agents to accept, reject, inspect, or ask about potentially incorrect user corrections — and introduces GAVA, which uses observation-bounded evidence, legal probes, and a one-step expected-loss rule.
- Evaluated in a text-only ALFWorld benchmark across many checkpoints and scenarios, GAVA achieves near-perfect correction accuracy while lowering interaction cost via selective inspection informed by an object-location prior, but the study is limited to symbolic observations, controlled speakers, and no human, visual, or robotic evaluation.
- This work shows selective information gathering with semantic priors can let interactive agents balance trust and verification cost, which matters for designing more efficient and reliable human–agent interaction.
WHY IT MATTERS
This work shows selective information gathering with semantic priors can let interactive agents balance trust and verification cost, which matters for designing more efficient and reliable human–agent interaction.