Academic Introduction Generation on ACL main conference papers 2025
0.98Semantic Similarity (SS)ELABORATE Prompt
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| ELABORATE PromptLLM=Llama3.1-8B-Instruct2025.12 | 0.98 | 0.8 | 0.447 | 19.981 | 1 | |
| STIGLLM=Llama3.1-8B-Instruct2025.12 | 0.978 | 0.836 | 0.472 | 20.717 | 1 | |
| STIGLLM=Qwen2.5-7B-Instruct2025.12 | 0.977 | 0.832 | 0.442 | 24.81 | 1 | |
| Pure PromptLLM=Qwen2.5-7B-Instruct2025.12 | 0.975 | 0.749 | 0.394 | 25.426 | 0.95 | |
| Pure PromptLLM=Llama3.1-8B-Instruct2025.12 | 0.975 | 0.772 | 0.427 | 14.843 | 1 | |
| GPTLLM=GPT-4o2025.12 | 0.973 | 0.779 | 0.399 | 28.233 | 1 | |
| ELABORATE PromptLLM=Qwen2.5-7B-Instruct2025.12 | 0.972 | 0.722 | 0.327 | 28.461 | 0.97 | |
| Outline WritingLLM=Qwen2.5-7B-Instruct2025.12 | 0.972 | 0.706 | 0.357 | 28.613 | 0.99 | |
| GPT (Elaborate)LLM=GPT-4o2025.12 | 0.971 | 0.903 | 0.496 | 31.906 | 1 | |
| AutoSurveyLLM=Qwen2.5-7B-Instruct2025.12 | 0.966 | 0.658 | 0.333 | 18.084 | 0.92 | |
| Outline WritingLLM=Llama3.1-8B-Instruct2025.12 | 0.958 | 0.759 | 0.404 | 26.328 | 1 |