Text-to-SQL on Text-to-SQL Multi-sharded
60.7Functional AccuracyRPOTD+replay
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| RPOTD+replayLLM of the system agent=Gemini2025.10 | 60.7 | 47.7 | 54.2 | |
| RPOTDLLM of the system agent=Gemini2025.10 | 56.1 | 40.8 | 28.9 | |
| MC-styleLLM of the system agent=Gemini2025.10 | 55.1 | 38.8 | 20.4 | |
| RPOTD+replayLLM of the system agent=GPT2025.10 | 52.8 | 47.7 | 54.2 | |
| APOLLM of the system agent=Gemini2025.10 | 52.3 | 36.8 | 16.9 | |
| GPOLLM of the system agent=Gemini2025.10 | 52.3 | 37.6 | 17.5 | |
| BaselineShardedLLM of the system agent=Gemini2025.10 | 51.4 | 33.3 | — | |
| RPOTD+replayLLM of the system agent=Llama-8B2025.10 | 46.7 | 47.7 | 54.2 | |
| MC-styleLLM of the system agent=GPT2025.10 | 45.9 | 38.8 | 20.4 | |
| GPOLLM of the system agent=GPT2025.10 | 45.8 | 37.6 | 17.5 | |
| RPOTDLLM of the system agent=GPT2025.10 | 43.9 | 40.8 | 28.9 | |
| BaselineShardedLLM of the system agent=GPT2025.10 | 40.2 | 33.3 | — | |
| RPOTD+replayLLM of the system agent=Llama-70B2025.10 | 40.2 | 47.7 | 54.2 | |
| BPOLLM of the system agent=Gemini2025.10 | 39.7 | 27.2 | 15.7 | |
| RPOTDLLM of the system agent=Llama-70B2025.10 | 38.3 | 40.8 | 28.9 | |
| RPOTD+replayLLM of the system agent=Llama-42025.10 | 38.3 | 47.7 | 54.2 | |
| APOLLM of the system agent=GPT2025.10 | 37.4 | 36.8 | 16.9 | |
| MC-styleLLM of the system agent=Llama-8B2025.10 | 34.6 | 38.8 | 20.4 | |
| APOLLM of the system agent=Llama-70B2025.10 | 33.6 | 36.8 | 16.9 | |
| RPOTDLLM of the system agent=Llama-42025.10 | 33.6 | 40.8 | 28.9 | |
| MC-styleLLM of the system agent=Llama-70B2025.10 | 33.2 | 38.8 | 20.4 | |
| BaselineShardedLLM of the system agent=Llama-70B2025.10 | 31.8 | 33.3 | — | |
| BPOLLM of the system agent=GPT2025.10 | 31.8 | 27.2 | 15.7 | |
| APOLLM of the system agent=Llama-42025.10 | 31.8 | 36.8 | 16.9 | |
| RPOTDLLM of the system agent=Llama-8B2025.10 | 31.8 | 40.8 | 28.9 | |
| GPOLLM of the system agent=Llama-70B2025.10 | 30.8 | 37.6 | 17.5 | |
| GPOLLM of the system agent=Llama-42025.10 | 29.9 | 37.6 | 17.5 | |
| APOLLM of the system agent=Llama-8B2025.10 | 29 | 36.8 | 16.9 | |
| GPOLLM of the system agent=Llama-8B2025.10 | 29 | 37.6 | 17.5 | |
| MC-styleLLM of the system agent=Llama-42025.10 | 25 | 38.8 | 20.4 | |
| BPOLLM of the system agent=Llama-70B2025.10 | 24.1 | 27.2 | 15.7 | |
| BaselineShardedLLM of the system agent=Llama-8B2025.10 | 22.4 | 33.3 | — | |
| BPOLLM of the system agent=Llama-42025.10 | 22 | 27.2 | 15.7 | |
| BaselineShardedLLM of the system agent=Llama-42025.10 | 20.6 | 33.3 | — | |
| BPOLLM of the system agent=Llama-8B2025.10 | 18.5 | 27.2 | 15.7 |