Trajectory Planning 벤치마크
Trajectory Planning on ToolBench
Win rate
- 2023-07-31 — GPT4- DFSDT: Win rate 70.4
- 2023-12-07 — Attention Bucket: Win rate 71.5
- 2024-02-15 — GPT4-TOPGUN: Win rate 86.54
| Rank | Model | Win rate | Paper | Code | Year |
|---|---|---|---|---|---|
| 1 | GPT4-TOPGUN | 86.54 | SwissNYF: Tool Grounded LLM Agents for Black Box Setting | iclr-dummy-user/swissnyf | 2024 |
| 2 | Attention Bucket | 71.5 | Fortify the Shortest Stave in Attention: Enhancing Context Awareness of Large Language Models for Effective Tool Use | fiorina1212/attention-buckets | 2023 |
| 3 | GPT4- DFSDT | 70.4 | ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs | openbmb/toolbench · eachsheep/shortcutsbench | 2023 |
| 4 | GPT4-TOPGUN | 86.54 | SwissNYF: Tool Grounded LLM Agents for Black Box Setting | iclr-dummy-user/swissnyf | 2024 |
| 5 | Attention Bucket | 71.5 | Fortify the Shortest Stave in Attention: Enhancing Context Awareness of Large Language Models for Effective Tool Use | fiorina1212/attention-buckets | 2023 |
| 6 | GPT4- DFSDT | 70.4 | ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs | openbmb/toolbench · eachsheep/shortcutsbench | 2023 |