docs: add Avg column to encoder comparison

README.md

@@ -138,21 +138,21 @@ Trained for 3 epochs (early-stopped) with a cosine schedule; training loss decre
 | Accuracy@1 | 0.5950 | 0.6480 | +0.053 (+8.9%) |
 | MAP@10 | 0.6013 | 0.6410 | +0.040 (+6.6%) |
 
-**v4 is the released model.** Fine-tuning lifts in-domain retrieval by **~6%** over the base `Qwen3-Embedding-4B`, with the largest gains on top-rank precision (Accuracy@1, Recall@1). It also surpasses the 0.6B sibling (test NDCG@10 0.6695) by **+0.021 (+3.2%)** — a modest scale gain at ~7× the parameters, so the 0.6B remains the better pick for latency-sensitive serving.
+**v4 is the released model.** Fine-tuning lifts in-domain retrieval by **roughly +7%** over the base `Qwen3-Embedding-4B`, with the largest gains on top-rank precision (Accuracy@1, Recall@1). It also surpasses the 0.6B sibling (test NDCG@10 0.6695) by **+0.021 (+3.2%)** — a modest scale gain at ~7× the parameters, so the 0.6B remains the better pick for latency-sensitive serving.
 
 ### Comparison with other encoders
 On the *same* in-domain test set, untuned encoders — our own `Qwen3-Embedding` base (0.6B / 4B) and public multilingual SOTA models (each run with its own native prompt format) — all fall **well below this model**: domain fine-tuning beats general-purpose scale:
 
-| Model | Params | NDCG@10 | MRR@10 | Recall@10 | Accuracy@1 | MAP@10 |
-|-------|:---:|:---:|:---:|:---:|:---:|:---:|
-| LiquidAI/LFM2.5-Embedding-350M | 0.35B | 0.5983 | 0.6166 | 0.6799 | 0.5320 | 0.5519 |
-| Qwen3-Embedding-0.6B (base) | 0.6B | 0.6186 | 0.6449 | 0.7046 | 0.5560 | 0.5652 |
-| google/embeddinggemma-300m | 0.3B | 0.6373 | 0.6664 | 0.7082 | 0.5790 | 0.5906 |
-| BAAI/bge-m3 | 0.6B | 0.6426 | 0.6660 | 0.7261 | 0.5730 | 0.5913 |
-| intfloat/multilingual-e5-large | 0.6B | 0.6476 | 0.6722 | 0.7313 | 0.5790 | 0.5958 |
-| Qwen3-Embedding-4B (base) | 4B | 0.6508 | 0.6805 | 0.7244 | 0.5950 | 0.6013 |
-| MoAI-Embedding-0.6B (sibling) | 0.6B | 0.6695 | 0.7060 | 0.7508 | 0.6190 | 0.6171 |
-| **MoAI-Embedding-4B (this model)** | 4B | **0.6906** | **0.7283** | **0.7620** | **0.6480** | **0.6410** |
+| Model | Params | NDCG@10 | MRR@10 | Recall@10 | Accuracy@1 | MAP@10 | Avg |
+|-------|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
+| LiquidAI/LFM2.5-Embedding-350M | 0.35B | 0.5983 | 0.6166 | 0.6799 | 0.5320 | 0.5519 | 0.5957 |
+| Qwen3-Embedding-0.6B (base) | 0.6B | 0.6186 | 0.6449 | 0.7046 | 0.5560 | 0.5652 | 0.6179 |
+| google/embeddinggemma-300m | 0.3B | 0.6373 | 0.6664 | 0.7082 | 0.5790 | 0.5906 | 0.6363 |
+| BAAI/bge-m3 | 0.6B | 0.6426 | 0.6660 | 0.7261 | 0.5730 | 0.5913 | 0.6398 |
+| intfloat/multilingual-e5-large | 0.6B | 0.6476 | 0.6722 | 0.7313 | 0.5790 | 0.5958 | 0.6452 |
+| Qwen3-Embedding-4B (base) | 4B | 0.6508 | 0.6805 | 0.7244 | 0.5950 | 0.6013 | 0.6504 |
+| MoAI-Embedding-0.6B (sibling) | 0.6B | 0.6695 | 0.7060 | 0.7508 | 0.6190 | 0.6171 | 0.6725 |
+| **MoAI-Embedding-4B (this model)** | 4B | **0.6906** | **0.7283** | **0.7620** | **0.6480** | **0.6410** | **0.6940** |
 
 This model improves over its own `Qwen3-Embedding-4B` base by **+0.040 NDCG@10 (+6.1%)** and leads the best general-purpose baseline (e5-large) by **+0.043 NDCG@10**. Notably, the untuned **4B base (`0.6508`) trails the fine-tuned 0.6B sibling (`0.6695`)** — fine-tuning outweighs scale. _Caveat: these baselines are not tuned on BC Card data — the comparison illustrates the value of domain adaptation, not a defect in the baselines._