Update README.md

README.md

@@ -125,6 +125,10 @@ similarities = query_emb.matmul(ctx_emb.transpose(0, 1)) # (1, num_ctx)
 ranked_results = torch.argsort(similarities, dim=-1, descending=True) # (1, num_ctx)
 ```
 
+## Evaluations on Multi-Turn QA Retrieval Benchmark
+**(UPDATE!!)** We evaluate multi-turn QA retrieval on five datasets: Doc2Dial, QuAC, QReCC, TopiOCQA, and INSCIT, which can be found in the [ChatRAG Bench](https://huggingface.co/datasets/nvidia/ChatRAG-Bench). The evaluation scripts can be found [here](https://huggingface.co/nvidia/dragon-multiturn-query-encoder/tree/main/evaluation).
+
+
 ## License
 Dragon-multiturn is built on top of [Dragon](https://arxiv.org/abs/2302.07452). We refer users to the original license of the Dragon model. Dragon-multiturn is also subject to the [Terms of Use](https://openai.com/policies/terms-of-use).