SEA is a diagnostic agent with a cognitively inspired dual-memory module and reinforcement training that jointly optimizes reasoning and memory, reaching 92.46% accuracy on MedCaseReasoning (+19.6% over baseline) and superior long-horizon gains on ER-Reason while producing clinically validated rules
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Joint Optimization of Reasoning and Dual-Memory for Self-Learning Diagnostic Agent
SEA is a diagnostic agent with a cognitively inspired dual-memory module and reinforcement training that jointly optimizes reasoning and memory, reaching 92.46% accuracy on MedCaseReasoning (+19.6% over baseline) and superior long-horizon gains on ER-Reason while producing clinically validated rules