Next Utterance on CSID (test)
68LLM-as-a-Judge ScoreSCRIPTMIND-FT
Evaluation results
| Method | Links | |
|---|---|---|
| SCRIPTMIND-FTbackbone=EEVE-Korean-Instruct-10.8B, fine-tuned=true2026.01 | 68 | |
| SCRIPTMIND-FTbackbone=EXAONE-3.5-2.4B-Instruct, fine-tuned=true2026.01 | 53 | |
| chatgpt-4o-latestzero-shot=true2026.01 | 45 | |
| SCRIPTMIND-FTbackbone=SOLAR-10.7B-Instruct, fine-tuned=true2026.01 | 44 | |
| EEVE-Korean-Instruct-10.8Bzero-shot=true2026.01 | 42 | |
| SCRIPTMIND-FTbackbone=Llama-3.1-8B-Instruct, fine-tuned=true2026.01 | 41 | |
| SCRIPTMIND-FTbackbone=Llama-3.2-1B-Instruct, fine-tuned=true2026.01 | 39 | |
| gemini-2.0-flashzero-shot=true2026.01 | 39 | |
| SCRIPTMIND-FTbackbone=Midm-2.0-Mini-Instruct, fine-tuned=true2026.01 | 33 | |
| SCRIPTMIND-FTbackbone=Midm-2.0-Base-Instruct, fine-tuned=true2026.01 | 33 | |
| claude-3-5-haikuzero-shot=true2026.01 | 31 | |
| EXAONE-3.5-2.4B-Instructzero-shot=true2026.01 | 30 | |
| EXAONE-3.0-7.8B-Instructzero-shot=true2026.01 | 26 | |
| Midm-2.0-Base-Instructzero-shot=true2026.01 | 22 | |
| Llama-3.1-8B-Instructzero-shot=true2026.01 | 19 | |
| SOLAR-10.7B-Instructzero-shot=true2026.01 | 16 | |
| Midm-2.0-Mini-Instructzero-shot=true2026.01 | 11 | |
| Llama-3.2-1B-Instructzero-shot=true2026.01 | 4 |