Task-oriented dialogue on FnCTOD
62.5Success RateRPOTD+replay
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| RPOTD+replayLLM of the meta-prompting agent=GPT2025.10 | 62.5 | — | — | |
| RPOTD+replayLLM of the meta-prompting agent=Gemini2025.10 | 62.2 | — | — | |
| RPOTD+replayLLM of the meta-prompting agent=Llama-8B2025.10 | 62.2 | — | — | |
| RPOTD+replayLLM of the meta-prompting agent=Llama-42025.10 | 61.8 | — | — | |
| RPOTD+replayLLM of the meta-prompting agent=Llama-70B2025.10 | 60.6 | — | — | |
| RPOTDLLM of the meta-prompting agent=Llama-8B2025.10 | 59.4 | — | — | |
| RPOTDLLM of the meta-prompting agent=Llama-42025.10 | 58.6 | — | — | |
| GPOLLM of the meta-prompting agent=GPT2025.10 | 57.9 | — | — | |
| RPOTDLLM of the meta-prompting agent=GPT2025.10 | 57.8 | — | — | |
| MC-styleLLM of the meta-prompting agent=Llama-42025.10 | 57.5 | — | — | |
| MC-styleLLM of the meta-prompting agent=Llama-70B2025.10 | 57.2 | — | — | |
| GPOLLM of the meta-prompting agent=Llama-42025.10 | 57.1 | — | — | |
| MC-styleLLM of the meta-prompting agent=GPT2025.10 | 56.7 | — | — | |
| RPOTDLLM of the meta-prompting agent=Gemini2025.10 | 56.2 | — | — | |
| APOLLM of the meta-prompting agent=Gemini2025.10 | 56 | — | — | |
| APOLLM of the meta-prompting agent=Llama-8B2025.10 | 56 | — | — | |
| APOLLM of the meta-prompting agent=Llama-70B2025.10 | 56 | — | — | |
| MC-styleLLM of the meta-prompting agent=Llama-8B2025.10 | 56 | — | — | |
| RPOTDLLM of the meta-prompting agent=Llama-70B2025.10 | 55.6 | — | — | |
| GPOLLM of the meta-prompting agent=Llama-8B2025.10 | 55.4 | — | — | |
| MC-styleLLM of the meta-prompting agent=Gemini2025.10 | 54.9 | — | — | |
| GPOLLM of the meta-prompting agent=Gemini2025.10 | 54.1 | — | — | |
| APOLLM of the meta-prompting agent=GPT2025.10 | 54 | — | — | |
| APOLLM of the meta-prompting agent=Llama-42025.10 | 54 | — | — | |
| GPOLLM of the meta-prompting agent=Llama-70B2025.10 | 52.6 | — | — | |
| APOAggregate=true2025.10 | — | 55.2 | 31.4 | |
| GPOAggregate=true2025.10 | — | 55.4 | 32 | |
| MC-styleAggregate=true2025.10 | — | 56.5 | 34.4 | |
| RPOTDAggregate=true2025.10 | — | 57.5 | 37 | |
| RPOTD+replayAggregate=true2025.10 | — | 61.9 | 47.3 |