Safetensors
Eval Results (legacy)
imbue2025 commited on
Commit
d89999f
·
verified ·
1 Parent(s): ffbd599

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +47 -1
README.md CHANGED
@@ -10,7 +10,6 @@ language:
10
 
11
  ![Model Architecture](https://img.shields.io/badge/Model-Socrates--embedding-blue) ![Parameter Count](https://img.shields.io/badge/Params-86M-green) ![Carbon Footprint](https://img.shields.io/badge/Carbon-Neutral-brightgreen)
12
 
13
-
14
  ## Model Details
15
 
16
  Socrates-embedding is a lightweight, high-density text embedding model. Unlike contemporary models that rely on massive parameter counts to brute-force semantic understanding, Socrates-embedding leverages Low-Rank Decay (LoRD) to achieve high-quality vector representations with minimal computational overhead.
@@ -21,6 +20,53 @@ This model is part of the Chunjiang Intelligence edge-computing initiative, aimi
21
  - **Model Type:** Dual-Encoder Transformer.
22
  - **Language:** English.
23
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
24
  ## Model Architecture
25
 
26
  The model utilizes a custom Transformer Encoder architecture optimized for inference latency on Apple MPS and NVIDIA TensorRT backends.
 
10
 
11
  ![Model Architecture](https://img.shields.io/badge/Model-Socrates--embedding-blue) ![Parameter Count](https://img.shields.io/badge/Params-86M-green) ![Carbon Footprint](https://img.shields.io/badge/Carbon-Neutral-brightgreen)
12
 
 
13
  ## Model Details
14
 
15
  Socrates-embedding is a lightweight, high-density text embedding model. Unlike contemporary models that rely on massive parameter counts to brute-force semantic understanding, Socrates-embedding leverages Low-Rank Decay (LoRD) to achieve high-quality vector representations with minimal computational overhead.
 
20
  - **Model Type:** Dual-Encoder Transformer.
21
  - **Language:** English.
22
 
23
+
24
+ The model was evaluated on the `AmazonCounterfactualClassification` dataset across multiple languages.
25
+
26
+ | Language | Accuracy |
27
+ | :--- | :---: |
28
+ | Japanese (ja) | 54.83 |
29
+ | German (de) | 52.57 |
30
+ | English (en) | 49.70 |
31
+ | English-Ext (en-ext)| 49.15 |
32
+
33
+ To put the model's efficiency into perspective, we compare its single-task score on Japanese classification against the *overall MTEB average scores* of much larger models. (Our budget is insufficient to cover the bill for the GPU used for the ongoing tests.)
34
+
35
+ <br>
36
+
37
+ <p align="center">
38
+ <img src="model_efficiency_comparison.png" width="800">
39
+ <br>
40
+ <em>Figure 1: Our 83M model's score on a single challenging task rivals the average performance of models up to 85x larger.</em>
41
+ </p>
42
+
43
+ <br>
44
+
45
+ Clustering performance was evaluated using the V-measure score (multiplied by 100) on the `StackExchangeClustering` task.
46
+
47
+ We compared Socrates-embedding against other popular lightweight models (<110M params).
48
+
49
+ | Model | Parameters | Clustering Score (V-measure x 100) |
50
+ | :--- | :--- | :---: |
51
+ | **Socrates-embedding** | **83M** | **8.92** 🏆 |
52
+ | `snowflake-arctic-embed-m`| 109M | 7.25 |
53
+ | `KartonBERT-USE-base-v1` | 104M | 6.93 |
54
+ | `jina-embedding-s-en-v1`| 35M | 6.64 |
55
+
56
+ | `all-MiniLM-L6-v2` | 23M | 6.62 |
57
+
58
+ * Observation: Our model achieves the highest clustering score in its weight class, demonstrating a superior vector space structure compared to established baselines.
59
+
60
+ <br>
61
+
62
+ <p align="center">
63
+ <img src="model_clustering_comparison.png" width="800">
64
+ <br>
65
+ <em>Figure 2: Leading clustering performance among lightweight embedding models.</em>
66
+ </p>
67
+
68
+ <br>
69
+
70
  ## Model Architecture
71
 
72
  The model utilizes a custom Transformer Encoder architecture optimized for inference latency on Apple MPS and NVIDIA TensorRT backends.