Update README.md

README.md

@@ -29,6 +29,8 @@ configs:
   data_files: single_turn/summarization/*.jsonl
 ---
 
+# LoopServe: An Adaptive Dual-phase LLM Inference Acceleration System for Multi-Turn Dialogues
+
 ``` shell
 .
 ├── README.md
@@ -45,5 +47,6 @@ configs:
     └── summarization
 ```
 
-# LoopServe: An Adaptive Dual-phase LLM Inference Acceleration System for Multi-Turn Dialogues
+# Introduction
 
+**LoopServe Multi-Turn Dialogue Benchmark** is a comprehensive evaluation dataset comprising 11 diverse datasets designed to assess large language model performance in realistic conversational scenarios. Unlike traditional benchmarks that place queries only at the end of input sequences, this benchmark features diverse query positions (beginning, middle, end) across 2-3 conversational turns, with datasets spanning Question Answering, Summarization, and Few-shot Learning tasks. The benchmark contains 199-500 instances per dataset with token lengths ranging from 2,376 to 30,545, capturing the dynamic dependencies and unpredictable patterns characteristic of real-world multi-turn dialogues to provide more authentic LLM evaluation.