Update README.md
Browse files
README.md
CHANGED
|
@@ -72,8 +72,11 @@ Extensive experiments on BEIR, MIRACL, and LMEB show that the KaLM-Reranker-V1 s
|
|
| 72 |
|
| 73 |
# Evaluation
|
| 74 |
## BEIR
|
|
|
|
| 75 |
|
| 76 |
## MIRACL
|
|
|
|
| 77 |
|
| 78 |
## LMEB
|
|
|
|
| 79 |
|
|
|
|
| 72 |
|
| 73 |
# Evaluation
|
| 74 |
## BEIR
|
| 75 |
+
On BEIR, KaLM-Reranker-V1 achieves state-of-the-art performance, on par with strong industrial models such as the Qwen3-Reranker series.
|
| 76 |
|
| 77 |
## MIRACL
|
| 78 |
+
On MIRACL, despite not being extensively trained on multilingual data, KaLM-Reranker-V1 still shows excellent reranking performance.
|
| 79 |
|
| 80 |
## LMEB
|
| 81 |
+
On LMEB, reranking models demonstrate a clear advantage, with even the 0.27B Nano model remaining competitive with 7–12B embedding models.
|
| 82 |
|