diff --git a/README.md b/README.md index 447f46d27a2c29e44353e720c3798e9fd7c19a0b..b23829f47dd5edaf6d49307459fae121af9a579b 100644 --- a/README.md +++ b/README.md @@ -2,6 +2,10 @@ ## Code for paper "G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment" [https://arxiv.org/abs/2303.16634] +## Model Evaluations + +Score prediction via gemini_eval.py, llama3_eval.py, qwen_eval.py + ## Experiments on SummEval dataset Full dataset used by G-Eval paper @@ -15,7 +19,7 @@ Sample dataset used for CoT analysis ## Prompts and Evaluation Results Prompts used to evaluate SummEval with GPT-4 & base and detailed prompts for CoT analysis are in prompts/summeval (by G-Eval paper) -Auto-CoT prompts are in prompts/cot_analysis +Auto-CoT prompts are in prompts/cot_analysis (created via auto_cot.py) GPT-4 G-eval results on SummEval are in results (by G-Eval paper) Other models results are in their respective folder \ No newline at end of file diff --git a/requirements.txt b/requirements.txt new file mode 100644 index 0000000000000000000000000000000000000000..a1e8c712bedd604302a4d8c50a66efb781b8b10d --- /dev/null +++ b/requirements.txt @@ -0,0 +1,6 @@ +groq==0.20.0 +openai==1.69.0 +prettytable==3.16.0 +protobuf==6.30.2 +scipy==1.15.2 +tqdm==4.67.1