Answer Generation on TopiOCQA
40.3F1 ScoreOurs
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| OursLLM=Qwen-2.5-7b2026.01 | 40.3 | 23.3 | |
| Search-R1LLM=Qwen-2.5-7b2026.01 | 37 | 20.7 | |
| AgenticLMLLM=Qwen-3-8b2026.01 | 34.6 | 15.3 | |
| ChatR1LLM=Qwen-2.5-7b2026.01 | 30.6 | — | |
| UniConvLLM=Mistral-2-7b2026.01 | 29.6 | — | |
| AgenticLMLLM=Qwen-3-4b2026.01 | 29.4 | 11.2 | |
| ChatR1LLM=Qwen-2.5-3b2026.01 | 29.4 | — | |
| REFRAGLLM=LLaMA-2-7b2026.01 | 28.2 | — | |
| ClaudeLLM=Sonnet-3.52026.01 | 27.2 | — | |
| EvoRAGLLM=Qwen-2.5-7b2026.01 | 26.8 | — | |
| Search-R1LLM=Qwen-2.5-3b2026.01 | 26.1 | 13.1 | |
| ChatGPTLLM=GPT-3.52026.01 | 25.5 | — | |
| OursLLM=Qwen-2.5-3b2026.01 | 25.2 | 14.1 | |
| SFTLLM=Qwen-2.5-7b2026.01 | 23.6 | 17.8 | |
| SFTLLM=Qwen-2.5-3b2026.01 | 18.2 | 14.7 | |
| ChatQALLM=LLaMA-3-8b2026.01 | 18.1 | — | |
| ConvSearch-R1LLM=Qwen-2.5-3b2026.01 | 7.6 | 3.5 |