Share your thoughts, 1 month free Claude Pro on usSee more

Sentiment Analysis on SST-2 (test)

97.1Accuracy

Previous SOTA

Updated 1mo ago

Evaluation Results

Method	Links
Previous SOTA 2020.05		97.1
RoBERTa + SKEP 2020.05		97
RoBERTa + SKEP 2020.05		96.7
LM-Cocktail2 2023.11		96.56
LM-Cocktail10 2023.11		96.56
RoBERTa 2020.05		96.5
CLS-based FT + TAPT 2023.05		96
PGI 2022.12		95.77
Non-Aligned 2024.02		95.6
Fine-tuned 2023.11		95.53
Prompt-based FT (hard) + PCP 2023.05		95.5
Prompt-based FT (hard) 2023.05		95.2
CLS-based FT 2023.05		95.1
Fine-tuning (full) 2021.08		95
Vaccine 2024.02		95
RoBERTa 2020.05		94.9
SFT 2024.02		94.8
Vlguard 2024.02		94.8
Vlguard 2024.02		94.8
Extra Annotation 2023.07		94.7
Non-Aligned 2024.02		94.6
Non-Aligned 2024.02		94.6
Vlguard 2024.02		94.6
Vlguard 2024.02		94.6
Vlguard 2024.02		94.6
GENICL 2025.05		94.6
CoTAM 2023.07		94.5
GENICL 2025.05		94.5
Non-Aligned 2024.02		94.4
SFT 2024.02		94.4
SFT 2024.02		94.4
Prompt-based FT (soft) + PCP 2023.05		94.3
FlipDA++ 2023.07		94.3
LLM-R 2025.05		94.3
LLM Pseudo Label 2023.07		94.2
SFT 2024.02		94.2
SFT 2024.02		94.2
Base 2023.07		94
COTDA 2023.07		94
Non-Aligned 2024.02		94
GENICL 2025.05		94
Prompt-based FT (soft) + PCP 2023.05		93.9
KEN 2024.02		93.8
Vaccine 2024.02		93.8
LLM-R 2025.05		93.8
Prompt-based FT (hard) + PCP 2023.05		93.6
Human Labeled 2022.12		93.52
DART 2021.08		93.5
Prompt-based FT (hard) + TAPT 2023.05		93.5
Bert-base 2024.02		93.37
Hybrid 2024.02		93.23
Vaccine 2024.02		93
LLM-R 2025.05		93
KEN 2024.02		92.9
GENICL 2025.05		92.7
Vaccine 2024.02		92.6
Vaccine 2024.02		92.6
LLM-R 2025.05		92.6
LKD+Init. 2023.05		92.5
LKD+Init. 2023.05		92.5
LM-BFF 2021.08		92.3
P-Tuning 2021.08		92.2
HybridNT 2024.02		92.2
LKD 2023.05		92
Hybrid 2024.02		91.97
ERM 2023.05		91.85
RETROPROMPT 2022.05		91.4
RETROPROMPT 2025.12		91.4
Gordon et al. (2020) 2024.02		90.8
HybridNT 2024.02		90.71
Q-Diversity 2023.05		90.62
LKD 2023.05		90.6
No Example 2023.07		90.5
Sajjad et al. (2020) 2024.02		90.3
Coherence Boosting (GPT-3 175B) 2021.10		89.84
LM-BFF (D-demo) 2022.05		89.3
LM-BFF (D-demo) 2025.12		89.3
PGDA 2022.12		89.29
LM-BFF (man) 2022.05		88.9
LM-BFF (man) 2025.12		88.9
EWC 2024.02		88.6
Dynamic Self-Attention 2018.08		88.5
Human Labeled 2022.12		88.47
CNN with self-attention 2018.08		88.3
BiLSTM with self-attention 2018.08		88.2
EWC 2024.02		88.2
GPT-3 175B 2021.10		88.14
PGDA 2022.12		87.75
BiLSTM 2018.08		87.5
EWC 2024.02		87.4
CNN 2018.08		87.2
Coherence Boosting (GPT-2 XL) 2021.10		86.93
KPT 2022.05		86.8
EWC 2024.02		86.8
KPT 2025.12		86.8
GPT-2 XL (1.6B) 2021.10		86.38
GPT-3 175B 2021.10		86.16
KL 2024.02		85.8
KL 2024.02		85.8
KL 2024.02		85.4

Showing 100 of 136 rows