Update README.md

README.md

@@ -21,6 +21,19 @@ tags:
 model-index:
 - name: CrossEncoder based on jhu-clsp/ettin-encoder-17m
   results:
+  - task:
+      type: semantic-similarity
+      name: Semantic Similarity
+    dataset:
+      name: sts test
+      type: sts-test
+    metrics:
+    - type: pearson_cosine
+      value: 0.8413715686076841
+      name: Pearson Cosine
+    - type: spearman_cosine
+      value: 0.8310895302151975
+      name: Spearman Cosine
   - task:
       type: semantic-similarity
       name: Semantic Similarity
@@ -28,12 +41,12 @@ model-index:
       name: sts dev
       type: sts-dev
     metrics:
-    - type: pearson
-      value: 0.8779482261036501
-      name: Pearson
-    - type: spearman
-      value: 0.8762065013920353
-      name: Spearman
+    - type: pearson_cosine
+      value: 0.8815197312565873
+      name: Pearson Cosine
+    - type: spearman_cosine
+      value: 0.8786002071426082
+      name: Spearman Cosine
 ---
 
 # EttinX Cross-Encoder: Semantic Similarity (STS)
@@ -49,7 +62,7 @@ which can easily process a few hundred sentence pairs per second on CPU, and a f
 ---
 
 ## Features
-- **High performing:** Achieves **Pearson: 0.8316** and **Spearman: 0.8211** on the STS-Benchmark test set.
+- **High performing:** Achieves **Pearson: 0.8414** and **Spearman: 0.8311** on the STS-Benchmark test set.
 - **Efficient architecture:** Based on the Ettin-encoder design (17M parameters), offering very fast inference speeds.
 - **Extended context length:** Processes sequences up to 8192 tokens, great for LLM output evals.
 - **Diversified training:** Pretrained on `dleemiller/wiki-sim` and fine-tuned on `sentence-transformers/stsb`.
@@ -64,7 +77,7 @@ which can easily process a few hundred sentence pairs per second on CPU, and a f
 | `ModernCE-base-sts`            | **0.9162**         | **0.9122**          | **8192**       | 149M       | **Fast** |
 | `stsb-roberta-large`           | 0.9147            | -              | 512            | 355M       | Slow    |
 | `stsb-distilroberta-base`      | 0.8792            | -              | 512            | 82M        | Fast    |
-| `EttinX-sts-xxs`               | 0.8316        | 0.8211          | **8192**       | 17M       | **Very Fast** |
+| `EttinX-sts-xxs`               | 0.8414        | 0.8311          | **8192**       | 17M       | **Very Fast** |
 
 
 ---
@@ -106,8 +119,8 @@ Fine-tuning was performed on the [`sentence-transformers/stsb`](https://huggingf
 
 ### Validation Results
 The model achieved the following test set performance after fine-tuning:
-- **Pearson Correlation:** 0.8316
-- **Spearman Correlation:** 0.8211
+- **Pearson Correlation:** 0.8414
+- **Spearman Correlation:** 0.8311
 
 ---