ResearchBenchmarksDialogue Question Answering on DialSim v1.1Follow49Overall F1 ScoreA-Mem41.7243.6145.547.39Apr 20, 2026Evaluation ResultsMethodMethodLinksOverall F1 ScoreAverage KMC Prompt Tokens (M)MC Time (h)QA Prompt Tokens (M)QA Time (h)A-MemMC/Retrieval Model=GPT...MC/Retrieval Model=GPT-4o-mini, Answer Generation Model=GPT-52026.044995.89.39.136.72.9HiGMemMC/Retrieval Model=GPT...MC/Retrieval Model=GPT-4o-mini, Answer Generation Model=GPT-52026.04423.925.215.123.54