Time-series question answering on TSQA 1.0 (test)
97.5F1 ScoreOpenTSLM SoftPrompt (Llama3.2-1B)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| OpenTSLM SoftPrompt (Llama3.2-1B)Category=OpenTSLM SoftPrompt, Base Model=Llama3.2-1B2025.10 | 97.5 | 97.54 | |
| OpenTSLM SoftPrompt (Llama3.2-3B)Category=OpenTSLM SoftPrompt, Base Model=Llama3.2-3B2025.10 | 97.37 | 97.33 | |
| Gemma3-4B FTCategory=Image (Plot), Protocol=Finetuned2025.10 | 94.94 | 94.85 | |
| OpenTSLM Flamingo (Llama3.2-1B)Category=OpenTSLM Flamingo, Base Model=Llama3.2-1B2025.10 | 94.08 | 94 | |
| OpenTSLM Flamingo (Gemma3-1B-pt)Category=OpenTSLM Flamingo, Base Model=Gemma3-1B-pt2025.10 | 92.56 | 92.46 | |
| OpenTSLM Flamingo (Llama3.2-3B)Category=OpenTSLM Flamingo, Base Model=Llama3.2-3B2025.10 | 90.14 | 90.1 | |
| OpenTSLM SoftPrompt (Gemma3-1B-pt)Category=OpenTSLM SoftPrompt, Base Model=Gemma3-1B-pt2025.10 | 87.29 | 89.18 | |
| Llama3.2-3BCategory=Tokenized Finetuned, Protocol=LoRA Finetuned2025.10 | 84.54 | 82.06 | |
| Llama3.2-1BCategory=Tokenized Finetuned, Protocol=LoRA Finetuned2025.10 | 83.74 | 81.4 | |
| Gemma3-1B-ptCategory=Tokenized Finetuned, Protocol=LoRA Finetuned2025.10 | 82.85 | 83.42 | |
| OpenTSLM Flamingo (Gemma3-270M)Category=OpenTSLM Flamingo, Base Model=Gemma3-270M2025.10 | 77.86 | 78.12 | |
| Gemma3-270MCategory=Tokenized Finetuned, Protocol=LoRA Finetuned2025.10 | 68.05 | 65.4 | |
| GPT-4oCategory=Image (Plot), Protocol=Zero-shot2025.10 | 59.24 | 62.1 | |
| Gemma3-4B-ptCategory=Image (Plot), Protocol=Zero-shot2025.10 | 48.77 | 50.6 | |
| GPT-4oCategory=Tokenized Time-Series, Protocol=Zero-shot2025.10 | 45.32 | 45.29 | |
| OpenTSLM SoftPrompt (Gemma3-270M)Category=OpenTSLM SoftPrompt, Base Model=Gemma3-270M2025.10 | 40.32 | 26.79 | |
| Random BaselineCategory=Random Baseline2025.10 | 33.33 | 33.33 | |
| Llama3.2-3BCategory=Tokenized Time-Series, Protocol=Zero-shot2025.10 | 16.24 | 32.06 | |
| Llama3.2-1BCategory=Tokenized Time-Series, Protocol=Zero-shot2025.10 | 16.01 | 31.04 | |
| Gemma3-1B-ptCategory=Tokenized Time-Series, Protocol=Zero-shot2025.10 | 11.76 | 12.92 | |
| Gemma3-270MCategory=Tokenized Time-Series, Protocol=Zero-shot2025.10 | 10.52 | 9.58 |