Update README.md

README.md

@@ -45,17 +45,24 @@ Surpassing Vicuna-7B and approximating LLaMA-2-Chat-7B on MT-Bench.
 
 **Instruction-following Benchmarks**
 
-|Method|AlpacaEval|MT-Bench|
-|--|--|--|
-|GPT-4|95.28|9.18|
-|Zephyr-7B-Beta|90.60|7.34|
-|Phi-2-DPO|81.37|-|
-|StableLM Zephyr 3B|76.00|6.64|
-|Vicuna-7B|76.84|6.17|
-|LLaMA-2-Chat-7B|71.37|6.27|
+|Method|AlpacaEval|MT-Bench|MT-Bench-ZH|
+|--|--|--|--|
+|GPT-4|95.28|9.18|8.96|
+|Zephyr-7B-Beta|90.60|7.34|6.27<sup>#</sup>|
+|Vicuna-7B|76.84|6.17|5.22<sup>#</sup>|
+|LLaMA-2-Chat-7B|71.37|6.27|5.43<sup>#</sup>|
+|Qwen-Chat-7B|-|-|6.24|
+|Phi-2-DPO|81.37|-|1.59<sup>#</sup><sup>$</sup>|
+|StableLM-Zephyr-3B|76.00|6.64|4.31<sup>#</sup>|
+|Rocket-3B|79.75|6.56|4.07<sup>#</sup>|
+|Qwen-Chat-1.8B|-|-|5.65|
 ||
-|MiniChat-3B|48.82|-|
-|MiniChat-2-3B|77.30|6.23|
+|MiniChat-3B|48.82|-|-|
+|MiniChat-2-3B|77.30|6.23|6.04|
+
+<sup>#</sup> specialized mainly for English.
+
+<sup>$</sup> finetuned without multi-turn instruction data.
 
 The following is an example code snippet to use MiniChat-2-3B: