General Question Answering on TriviaQA (EM, F1aware)
64.4EMSearch-R1
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Search-R1Base Model=Llama-3.2-3B-Instruct, Category=Outcome RL Baselines2025.12 | 64.4 | 2.5 | |
| ADASEARCHBase Model=Llama-3.2-3B-Instruct, Category=Paper Method2025.12 | 62.9 | 79.2 | |
| IKEABase Model=Llama-3.2-3B-Instruct, Category=Reward-Shaping Baselines2025.12 | 62 | — | |
| Awareness ShapingBase Model=Llama-3.2-3B-Instruct, Category=Reward-Shaping Baselines2025.12 | 61.5 | 75.4 | |
| Search-R1Base Model=Qwen2.5-3B-Instruct, Category=Outcome RL Baselines2025.12 | 60.1 | 0.3 | |
| Naive ShapingBase Model=Llama-3.2-3B-Instruct, Category=Reward-Shaping Baselines2025.12 | 59.7 | 75.3 | |
| IKEABase Model=Qwen2.5-3B-Instruct, Category=Reward-Shaping Baselines2025.12 | 59.5 | 62.4 | |
| ADASEARCHBase Model=Qwen2.5-3B-Instruct, Category=Paper Method2025.12 | 56.8 | 71.7 | |
| Awareness ShapingBase Model=Qwen2.5-3B-Instruct, Category=Reward-Shaping Baselines2025.12 | 56.1 | 70.3 | |
| Naive ShapingBase Model=Qwen2.5-3B-Instruct, Category=Reward-Shaping Baselines2025.12 | 54.8 | 66.9 | |
| RL w/o searchBase Model=Llama-3.2-3B-Instruct, Category=Outcome RL Baselines2025.12 | 54.6 | 54.6 | |
| RAGBase Model=Llama-3.2-3B-Instruct, Category=Prompting Baselines2025.12 | 53 | 0 | |
| RAGBase Model=Qwen2.5-3B-Instruct, Category=Prompting Baselines2025.12 | 52.7 | 0 | |
| CoTBase Model=Llama-3.2-3B-Instruct, Category=Prompting Baselines2025.12 | 45.6 | 45.6 | |
| RL w/o searchBase Model=Qwen2.5-3B-Instruct, Category=Outcome RL Baselines2025.12 | 44.9 | 44.9 | |
| Direct AnswerBase Model=Llama-3.2-3B-Instruct, Category=Prompting Baselines2025.12 | 42.3 | 42.3 | |
| CoTBase Model=Qwen2.5-3B-Instruct, Category=Prompting Baselines2025.12 | 38.2 | 38.2 | |
| Direct AnswerBase Model=Qwen2.5-3B-Instruct, Category=Prompting Baselines2025.12 | 24.1 | 24.1 |