Sentiment Analysis on DynaSent

Metric: Macro F1 (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

#	Model↕	Macro F1▼	Extra Data	Paper	Date↕	Code
1	GPT-4o Fine-Tuned (Minimal)	89	No	ELECTRA and GPT-4o: Cost-Effective Partners for ...	2024-12-29	Code
2	GPT-4o-mini Fine-Tuned	86.9	No	ELECTRA and GPT-4o: Cost-Effective Partners for ...	2024-12-29	Code
3	GPT-4o + ELECTRA Large FT (Prompt, Label, Examples)	81.53	No	ELECTRA and GPT-4o: Cost-Effective Partners for ...	2024-12-29	Code
4	GPT-4o (Prompt)	80.22	No	ELECTRA and GPT-4o: Cost-Effective Partners for ...	2024-12-29	Code
5	GPT-4o-mini + ELECTRA Large FT (Prompt, Label, Probabilities)	79.72	No	ELECTRA and GPT-4o: Cost-Effective Partners for ...	2024-12-29	Code
6	GPT-4o-mini + ELECTRA Large FT (Prompt, Label)	77.94	No	ELECTRA and GPT-4o: Cost-Effective Partners for ...	2024-12-29	Code
7	GPT-4o + ELECTRA Large FT	77.69	No	ELECTRA and GPT-4o: Cost-Effective Partners for ...	2024-12-29	Code
8	GPT-4o-mini (Prompt)	77.35	No	ELECTRA and GPT-4o: Cost-Effective Partners for ...	2024-12-29	Code
9	ELECTRA Large Fine-Tuned	76.29	No	ELECTRA and GPT-4o: Cost-Effective Partners for ...	2024-12-29	Code
10	GPT-4o-mini + ELECTRA Base FT	76.19	No	ELECTRA and GPT-4o: Cost-Effective Partners for ...	2024-12-29	Code
11	ELECTRA Base Fine-Tuned	71.83	No	ELECTRA and GPT-4o: Cost-Effective Partners for ...	2024-12-29	Code