Question Answering on MPVQA
0.1615ANLSMultiDocFusion
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| MultiDocFusiontop-k={1, 4}, Models=Llama-3.2-3B, Mistral-8B, Qwen-2.5-7B (averaged)2026.04 | 0.1615 | 0.1316 | 0.185 | |
| Structure-based chunkingtop-k={1, 4}, Models=Llama-3.2-3B, Mistral-8B, Qwen-2.5-7B (averaged)2026.04 | 0.1537 | 0.098 | 0.1278 | |
| Length chunkingtop-k={1, 4}, Models=Llama-3.2-3B, Mistral-8B, Qwen-2.5-7B (averaged)2026.04 | 0.1398 | 0.0966 | 0.1408 | |
| Perplexity chunkingtop-k={1, 4}, Models=Llama-3.2-3B, Mistral-8B, Qwen-2.5-7B (averaged)2026.04 | 0.1344 | 0.0751 | 0.095 | |
| Semantic chunkingtop-k={1, 4}, Models=Llama-3.2-3B, Mistral-8B, Qwen-2.5-7B (averaged)2026.04 | 0.1332 | 0.0805 | 0.0978 | |
| LumberChunkertop-k={1, 4}, Models=Llama-3.2-3B, Mistral-8B, Qwen-2.5-7B (averaged)2026.04 | 0.1307 | 0.0769 | 0.0993 |