Zero-shot Text Classification on B2MWT-10C (test)
75.49Macro F1Voyage-4-nano
Evaluation results
| Method | Links | |
|---|---|---|
| Voyage-4-nanoRefinement Method=History-Aware (M3), LLM=mistral, k=4, m=42026.04 | 75.49 | |
| Voyage-4-nanoRefinement Method=Example-Guided (M1), LLM=mistral2026.04 | 74.8 | |
| Voyage-4-nanoRefinement Method=Confusion-Aware (M2), LLM=mistral, k=32026.04 | 74.36 | |
| KaLM-Embedding-v2.5Refinement Method=Confusion-Aware (M2), LLM=mistral, k=12026.04 | 69.87 | |
| KaLM-Embedding-v2.5Refinement Method=Example-Guided (M1), LLM=mistral2026.04 | 68.62 | |
| KaLM-Embedding-v2.5Refinement Method=History-Aware (M3), LLM=mistral, k=5, m=12026.04 | 68.52 | |
| Qwen3-Embedding-0.6BRefinement Method=Example-Guided (M1), LLM=glm-42026.04 | 68.17 | |
| Qwen3-Embedding-0.6BRefinement Method=Confusion-Aware (M2), LLM=mistral, k=52026.04 | 68.06 | |
| Embeddinggemma-300mRefinement Method=History-Aware (M3), LLM=mistral, k=5, m=12026.04 | 67.65 | |
| Qwen3-Embedding-0.6BRefinement Method=History-Aware (M3), LLM=mistral, k=5, m=12026.04 | 67.54 | |
| Embeddinggemma-300mRefinement Method=Example-Guided (M1), LLM=qwen2026.04 | 66.39 | |
| Embeddinggemma-300mRefinement Method=Confusion-Aware (M2), LLM=gemma, k=42026.04 | 65.82 | |
| Modernbert-embed-largeRefinement Method=Example-Guided (M1), LLM=qwen2026.04 | 65.69 | |
| Modernbert-embed-largeRefinement Method=History-Aware (M3), LLM=glm-4, k=3, m=22026.04 | 64.65 | |
| Mxbai-embed-large-v1Refinement Method=History-Aware (M3), LLM=mistral, k=5, m=12026.04 | 63.5 | |
| Llama-nemotron-embed-1b-v2Refinement Method=Example-Guided (M1), LLM=qwen2026.04 | 62.68 | |
| Voyage-4-nanoRefinement Method=Baseline2026.04 | 62.65 | |
| Modernbert-embed-largeRefinement Method=Confusion-Aware (M2), LLM=mistral, k=32026.04 | 62.3 | |
| Mxbai-embed-large-v1Refinement Method=Example-Guided (M1), LLM=glm-42026.04 | 61.93 | |
| Snowflake-arctic-embed-v2.0Refinement Method=History-Aware (M3), LLM=mistral, k=5, m=12026.04 | 61.79 | |
| Snowflake-arctic-embed-v2.0Refinement Method=Example-Guided (M1), LLM=qwen2026.04 | 61.37 | |
| Mxbai-embed-large-v1Refinement Method=Confusion-Aware (M2), LLM=mistral, k=22026.04 | 61.06 | |
| Llama-nemotron-embed-1b-v2Refinement Method=History-Aware (M3), LLM=gemma, k=3, m=12026.04 | 60.82 | |
| BAAI bge-m3Refinement Method=History-Aware (M3), LLM=mistral, k=5, m=12026.04 | 60.6 | |
| Alibaba-NLP-gteRefinement Method=History-Aware (M3), LLM=mistral, k=5, m=12026.04 | 60.44 | |
| Llama-nemotron-embed-1b-v2Refinement Method=Confusion-Aware (M2), LLM=gemma, k=42026.04 | 60.29 | |
| BAAI bge-m3Refinement Method=Confusion-Aware (M2), LLM=mistral, k=22026.04 | 59.82 | |
| BAAI bge-m3Refinement Method=Example-Guided (M1), LLM=gemma2026.04 | 59.67 | |
| KaLM-Embedding-v2.5Refinement Method=Baseline2026.04 | 59.6 | |
| Snowflake-arctic-embed-v2.0Refinement Method=Confusion-Aware (M2), LLM=mistral, k=22026.04 | 59.59 | |
| Alibaba-NLP-gteRefinement Method=Confusion-Aware (M2), LLM=qwen, k=52026.04 | 59.25 | |
| Alibaba-NLP-gteRefinement Method=Example-Guided (M1), LLM=mistral2026.04 | 59.16 | |
| Solon-embeddings-large-0.1Refinement Method=History-Aware (M3), LLM=mistral, k=2, m=12026.04 | 58.38 | |
| Solon-embeddings-large-0.1Refinement Method=Example-Guided (M1), LLM=mistral2026.04 | 58.01 | |
| Solon-embeddings-large-0.1Refinement Method=Confusion-Aware (M2), LLM=mistral, k=12026.04 | 57.92 | |
| Granite-278m-multilingualRefinement Method=Example-Guided (M1), LLM=qwen2026.04 | 57.55 | |
| Granite-278m-multilingualRefinement Method=History-Aware (M3), LLM=mistral, k=4, m=52026.04 | 57 | |
| Qwen3-Embedding-0.6BRefinement Method=Baseline2026.04 | 56.15 | |
| Granite-278m-multilingualRefinement Method=Confusion-Aware (M2), LLM=qwen, k=22026.04 | 55.46 | |
| Modernbert-embed-largeRefinement Method=Baseline2026.04 | 55.02 | |
| Mxbai-embed-large-v1Refinement Method=Baseline2026.04 | 54.75 | |
| BAAI bge-m3Refinement Method=Baseline2026.04 | 52.75 | |
| Granite-278m-multilingualRefinement Method=Baseline2026.04 | 52.18 | |
| Alibaba-NLP-gteRefinement Method=Baseline2026.04 | 52.14 | |
| Snowflake-arctic-embed-v2.0Refinement Method=Baseline2026.04 | 50.1 | |
| Llama-nemotron-embed-1b-v2Refinement Method=Baseline2026.04 | 49.19 | |
| Multilingual-e5-largeRefinement Method=History-Aware (M3), LLM=mistral, k=4, m=42026.04 | 49.07 | |
| Embeddinggemma-300mRefinement Method=Baseline2026.04 | 48.64 | |
| Solon-embeddings-large-0.1Refinement Method=Baseline2026.04 | 48.12 | |
| Multilingual-e5-largeRefinement Method=Confusion-Aware (M2), LLM=mistral, k=32026.04 | 47.34 | |
| Multilingual-e5-largeRefinement Method=Example-Guided (M1), LLM=mistral2026.04 | 46.48 | |
| Multilingual-e5-largeRefinement Method=Baseline2026.04 | 41.77 |