Question Answering on DuReader (test)
23.87F1 ScoreMeta-chunker-1.5B
Evaluation results
| Method | Links | |
|---|---|---|
| Meta-chunker-1.5BChunking Method=Meta-chunker, Backbone=Qwen2.5-1.5B2025.03 | 23.87 | |
| Qwen2.5-72BChunking Method=Direct LLM Inference, Backbone=Qwen2.5-72B2025.03 | 22.84 | |
| Qwen2.5-14BChunking Method=Direct LLM Inference, Backbone=Qwen2.5-14B2025.03 | 22.71 | |
| Llama_indexChunking Method=LlamaIndex default2025.03 | 22.2 | |
| LumberChunkerChunking Method=Model-based (Qwen2.5-14B)2025.03 | 21.78 | |
| Semantic ChunkingChunking Method=Semantic-based2025.03 | 21.57 | |
| OriginalChunking Method=Rule-based/Original2025.03 | 20.3 |