Count Querying on CoP-QA-F
1AC ScoreTalk2DM
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Talk2DMLLM Backbone=Qwen3:30B2026.02 | 1 | 0.9905 | |
| Talk2DMLLM Backbone=Magistral:24B2026.02 | 0.9938 | 0.8021 | |
| Talk2DMLLM Backbone=Gemma3:27B2026.02 | 0.9918 | 0.9649 | |
| Talk2DMLLM Backbone=GPT-oss:20B2026.02 | 0.9907 | 0.9659 | |
| Talk2DMLLM Backbone=Llama3.1:8B2026.02 | 0.9856 | 0.899 | |
| Talk2DMLLM Backbone=Deepseek-r1:32B2026.02 | 0.8889 | 0.8788 |