Summarization on SummaryBench Dynasts
99.7Structural ScoreAPWA
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| APWAConfig=5.4×mini2026.05 | 99.7 | 45.1 | — | — | |
| DirectLLM Backend=GPT-5.4 mini2026.05 | 97.9 | 21 | — | — | |
| Direct2026.05 | 97.9 | 21 | — | — | |
| APWAConfig=mini×mini2026.05 | 97.1 | 44.7 | — | — | |
| APWALLM Backend=GPT-5.4 mini2026.05 | 95.4 | 41.9 | — | — | |
| APWAConfig=5.4×nano2026.05 | 92.3 | 41.9 | — | — | |
| APWAConfig=mini×nano2026.05 | 87.2 | 32.3 | — | — | |
| MegaAgentLLM Backend=GPT-4.1 mini2026.05 | 21.2 | 2.3 | — | — | |
| MegaAgent2026.05 | 21.2 | 2.3 | — | — | |
| APWALLM Backend=GPT-5.4 mini2026.05 | — | — | 0 | 210 | |
| DirectLLM Backend=GPT-5.4 mini2026.05 | — | — | 60 | 76 | |
| Magentic-OneLLM Backend=GPT-5.4 mini2026.05 | — | — | 100 | — | |
| MegaAgentLLM Backend=GPT-4.1 mini2026.05 | — | — | 80 | 248 |