Single-document Question Answering on Single-doc QA
62.3F1ChatGLM3-6B-32K
Evaluation results
| Method | Links | |
|---|---|---|
| ChatGLM3-6B-32KContext Window Size=32K, Backbone=6B2023.11 | 62.3 | |
| GPT3.5-Turbo-16KContext Window Size=16K2023.11 | 61.2 | |
| Baichuan2-Turbo-192KContext Window Size=192K2023.11 | 44.7 | |
| Vicuna-v1.5-7B-16KContext Window Size=16K, Backbone=7B2023.11 | 43 | |
| Our modelContext Window Size=8K2023.11 | 34.4 | |
| InternLM-7B-8KContext Window Size=8K, Backbone=7B2023.11 | 33.6 | |
| ChatGLM2-6B-32KContext Window Size=32K, Backbone=6B2023.11 | 32.8 | |
| Qwen-14B-ChatContext Window Size=8K, Backbone=14B2023.11 | 31.4 | |
| Longchat-v1.5-7B-32KContext Window Size=32K, Backbone=7B2023.11 | 29.1 | |
| Xgen-7B-8KContext Window Size=8K, Backbone=7B2023.11 | 14.8 |