Long-context Reasoning on ∞ Bench
90.39AccuracyMiA (Emb-Only)
Evaluation Results
| Method | Links | |
|---|---|---|
| MiA (Emb-Only)Retriever=MiA-Emb-8B, +S (Retriever)=true, Generator=Qwen2.5-72B, +S (Generator)=false, top-k=102025.12 | 90.39 | |
| MiARetriever=MiA-Emb-8B, +S (Retriever)=true, Generator=Qwen2.5-72B, +S (Generator)=true, top-k=102025.12 | 88.21 | |
| MiA (Emb-Only)Retriever=MiA-Emb-8B, +S (Retriever)=true, Generator=Qwen2.5-72B, +S (Generator)=false, top-k=52025.12 | 87.77 | |
| VanillaRetriever=Qwen3-8B, +S (Retriever)=false, Generator=Qwen2.5-72B, +S (Generator)=false, top-k=102025.12 | 86.9 | |
| MiA (Gen-Only)Retriever=Qwen3-8B, +S (Retriever)=false, Generator=Qwen2.5-72B, +S (Generator)=true, top-k=102025.12 | 86.46 | |
| MiARetriever=MiA-Emb-8B, +S (Retriever)=true, Generator=Qwen2.5-72B, +S (Generator)=true, top-k=52025.12 | 86.46 | |
| MiA-RAGRetriever=MiA-Emb-8B, +S (Retriever)=true, Generator=MiA-Gen-14B, +S (Generator)=true, top-k=102025.12 | 85.15 | |
| MiA (Emb-Only)Retriever=MiA-Emb-8B, +S (Retriever)=true, Generator=Qwen2.5-72B, +S (Generator)=false, top-k=32025.12 | 84.72 | |
| MiARetriever=MiA-Emb-8B, +S (Retriever)=true, Generator=Qwen2.5-72B, +S (Generator)=true, top-k=32025.12 | 84.71 | |
| MiA (Emb-Only)Retriever=MiA-Emb-8B, +S (Retriever)=true, Generator=Qwen2.5-14B, +S (Generator)=false, top-k=102025.12 | 84.28 | |
| MiARetriever=MiA-Emb-8B, +S (Retriever)=true, Generator=Qwen2.5-14B, +S (Generator)=true, top-k=102025.12 | 84.28 | |
| MiA (Gen-Only)Retriever=Qwen3-8B, +S (Retriever)=false, Generator=Qwen2.5-72B, +S (Generator)=true, top-k=52025.12 | 83.84 | |
| MiARetriever=MiA-Emb-8B, +S (Retriever)=true, Generator=Qwen2.5-14B, +S (Generator)=true, top-k=52025.12 | 83.41 | |
| MiA (Emb-Only)Retriever=MiA-Emb-8B, +S (Retriever)=true, Generator=Qwen2.5-14B, +S (Generator)=false, top-k=52025.12 | 82.97 | |
| MiA (Gen-Only)Retriever=Qwen3-8B, +S (Retriever)=false, Generator=Qwen2.5-72B, +S (Generator)=true, top-k=32025.12 | 82.1 | |
| MiA (Gen-Only)Retriever=Qwen3-8B, +S (Retriever)=false, Generator=MiA-Gen-14B, +S (Generator)=true, top-k=52025.12 | 82.1 | |
| MiA-RAGRetriever=MiA-Emb-8B, +S (Retriever)=true, Generator=MiA-Gen-14B, +S (Generator)=true, top-k=52025.12 | 81.22 | |
| VanillaRetriever=Qwen3-8B, +S (Retriever)=false, Generator=Qwen2.5-72B, +S (Generator)=false, top-k=52025.12 | 80.79 | |
| MiA (Gen-Only)Retriever=Qwen3-8B, +S (Retriever)=false, Generator=Qwen2.5-14B, +S (Generator)=true, top-k=52025.12 | 80.79 | |
| MiA (Gen-Only)Retriever=Qwen3-8B, +S (Retriever)=false, Generator=MiA-Gen-14B, +S (Generator)=true, top-k=102025.12 | 80.79 | |
| MiA-RAGRetriever=MiA-Emb-8B, +S (Retriever)=true, Generator=MiA-Gen-14B, +S (Generator)=true, top-k=32025.12 | 80.79 | |
| MiARetriever=MiA-Emb-8B, +S (Retriever)=true, Generator=Qwen2.5-14B, +S (Generator)=true, top-k=32025.12 | 79.91 | |
| MiA (Gen-Only)Retriever=Qwen3-8B, +S (Retriever)=false, Generator=Qwen2.5-14B, +S (Generator)=true, top-k=102025.12 | 79.48 | |
| MiA (Emb-Only)Retriever=MiA-Emb-8B, +S (Retriever)=true, Generator=Qwen2.5-14B, +S (Generator)=false, top-k=32025.12 | 79.48 | |
| VanillaRetriever=Qwen3-8B, +S (Retriever)=false, Generator=Qwen2.5-14B, +S (Generator)=false, top-k=102025.12 | 77.29 | |
| MiA (Gen-Only)Retriever=Qwen3-8B, +S (Retriever)=false, Generator=Qwen2.5-14B, +S (Generator)=true, top-k=32025.12 | 75.98 | |
| MiA (Gen-Only)Retriever=Qwen3-8B, +S (Retriever)=false, Generator=MiA-Gen-14B, +S (Generator)=true, top-k=32025.12 | 75.98 | |
| VanillaRetriever=Qwen3-8B, +S (Retriever)=false, Generator=Qwen2.5-72B, +S (Generator)=false, top-k=32025.12 | 75.55 | |
| VanillaRetriever=Qwen3-8B, +S (Retriever)=false, Generator=Qwen2.5-14B, +S (Generator)=false, top-k=52025.12 | 73.8 | |
| VanillaRetriever=Qwen3-8B, +S (Retriever)=false, Generator=Qwen2.5-14B, +S (Generator)=false, top-k=32025.12 | 72.49 | |
| Summary-OnlyGenerator=Qwen2.5-72B, +S (Generator)=true2025.12 | 72.05 | |
| Summary-OnlyGenerator=Qwen2.5-14B, +S (Generator)=true2025.12 | 61.57 |