Fact Consolidation (Single-Hop) on MemoryAgentBench (MAB) FC-SH 6K
99AccuracyAblation C (gpt-4o backbone)
Evaluation Results
| Method | Links | |
|---|---|---|
| Ablation C (gpt-4o backbone)Pipeline=Ablation C (gpt-4o backbone), Backbone=gpt-4o2026.05 | 99 | |
| Ablation A (chunk-4096)Pipeline=Ablation A (chunk-4096), Backbone=gpt-4o-mini2026.05 | 87 | |
| HeadlinePipeline=Headline, Backbone=gpt-4o-mini2026.05 | 71 | |
| LLM-judgment baselinePipeline=LLM-judgment baseline, Backbone=gpt-4o-mini2026.05 | 63 |