Question Answering on OBQA (Accuracy, Tokens)
95.2AccuracyPACT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| PACTBase Model=Qwen3-14B, Pipeline Setting=four-agent sequential pipeline (Setting B)2026.06 | 95.2 | 3,559 | |
| PACTBase Model=Qwen3-32B, Pipeline Setting=four-agent sequential pipeline (Setting B)2026.06 | 94.4 | 3,517 | |
| CoABase Model=Qwen3-14B, Pipeline Setting=four-agent sequential pipeline (Setting B)2026.06 | 93.9 | 4,533 | |
| TextMASBase Model=Qwen3-32B, Pipeline Setting=four-agent sequential pipeline (Setting B)2026.06 | 93.8 | 5,770 | |
| Multi-Agent DebateBase Model=Qwen3-32B, Pipeline Setting=four-agent sequential pipeline (Setting B)2026.06 | 93.7 | 28,394 | |
| Multi-Agent DebateBase Model=Qwen3-14B, Pipeline Setting=four-agent sequential pipeline (Setting B)2026.06 | 93.2 | 30,704 | |
| CoABase Model=Qwen3-32B, Pipeline Setting=four-agent sequential pipeline (Setting B)2026.06 | 93.1 | 3,811 | |
| PACTBase Model=Qwen3-8B, Pipeline Setting=four-agent sequential pipeline (Setting B)2026.06 | 92.5 | 4,304 | |
| Multi-Agent DebateBase Model=Qwen3-8B, Pipeline Setting=four-agent sequential pipeline (Setting B)2026.06 | 92 | 32,784 | |
| TextMASBase Model=Qwen3-14B, Pipeline Setting=four-agent sequential pipeline (Setting B)2026.06 | 92 | 4,675 | |
| TextMASBase Model=Qwen3-8B, Pipeline Setting=four-agent sequential pipeline (Setting B)2026.06 | 91.1 | 5,263 | |
| CoABase Model=Qwen3-8B, Pipeline Setting=four-agent sequential pipeline (Setting B)2026.06 | 90.4 | 4,851 |