Upload README.md

README.md

@@ -24,8 +24,9 @@ language:
 1. [Introduction](#introduction)
 2. [Technical Report](#technical-report)
 3. [Highlights](#highlights)
-4. [Quickstart](#quickstart)
-5. [Citation](#citation)
+4. [Benchmarks](#benchmark-comparison)
+5. [Quickstart](#quickstart)
+6. [Citation](#citation)
 
 # Introduction
 
@@ -48,12 +49,15 @@ A technical report detailing our proposed `LEAF` training procedure will be avai
 * **Flexible Architecture Support**: `mdbr-leaf-ir` supports asymmetric retrieval architectures enabling even greater retrieval results. [See below](#asymmetric-retrieval-setup) for more information.
 * **MRL and Quantization Support**: embedding vectors generated by `mdbr-leaf-ir` compress well when truncated (MRL) and can be stored using more efficient types like `int8` and `binary`.  [See below](#mrl-truncation) for more information.
 
-## Benchmark Comparison  
+## Benchmark Comparison
   
 The table below shows the average BEIR benchmark scores (nDCG@10) for `mdbr-leaf-ir` compared to other retrieval models.
 
+`mdbr-leaf-ir` ranks #1 on the BEIR public leaderboard, and when run in asymmetric "**(asym.)**" mode as described [here](#asymmetric-retrieval-setup), the results improve even further.
+
 | Model                              | Size | BEIR  Avg. (nDCG@10) |  
 |------------------------------------|------|----------------------|
+| **mdbr-leaf-ir (asym.)**           | 23M  | **54.03**            |  
 | **mdbr-leaf-ir**                   | 23M  | **53.55**            |  
 | snowflake-arctic-embed-s           | 32M  | 51.98                |  
 | bge-small-en-v1.5                  | 33M  | 51.65                |  
@@ -64,7 +68,7 @@ The table below shows the average BEIR benchmark scores (nDCG@10) for `mdbr-leaf
 | MiniLM-L6-v2                       | 23M  | 41.95                |  
 | BM25                               | –    | 41.14                |  
 
-[//]: # (| **mdbr-leaf-ir (asym.)**           | 23M  | **?**      |  )
+
 
 
 # Quickstart  
@@ -114,7 +118,7 @@ for i, query in enumerate(queries):
 
 See full example notebook [here](https://huggingface.co/MongoDB/mdbr-leaf-ir/blob/main/transformers_example.ipynb).
   
-## Asymmetric Retrieval Setup  
+## Asymmetric Retrieval Setup
   
 `mdbr-leaf-ir` is *aligned* to [`snowflake-arctic-embed-m-v1.5`](https://huggingface.co/Snowflake/snowflake-arctic-embed-m-v1.5), the model it has been distilled from. This enables flexible architectures in which, for example, documents are encoded using the larger model, while queries can be encoded faster and more efficiently with the compact `leaf` model:
 ```python