A latency-aware evaluation for HAZARD shows that a reflex-plus-async-LLM agent beats rule-based baselines in the fire scenario.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
LLM-Enhanced Rapid-Reflex Async-Reflect Embodied Agent for Real-Time Decision-Making in Dynamically Changing Environments
A latency-aware evaluation for HAZARD shows that a reflex-plus-async-LLM agent beats rule-based baselines in the fire scenario.