Our new X account is live! Follow @wizwand_team for updates

Social Interaction Question Answering on SIQA

86.9Accuracy

Human Performance

Updated 4d ago

Evaluation Results

Method	Links
Human Performance 2024.01		86.9
T5 2020.05		81.4
UNIFIEDQA 2020.05		81.4
DeBERTa-v3-L (Supervised) 2024.01		80.1
VERA-T5 (Multitask Supervised) 2024.01		80.1
RoBERTa 2020.05		78
ROBERTa-L (Supervised) 2024.01		76.6
BART-large 2020.05		74
UnifiedQA-BART 2020.05		73.2
ChatGPT + Chain-of-thought 2024.01		70.7
ChatGPT + Self-consistent chain-of-thought 2024.01		69.7
ChatGPT 2024.01		69.5
GPT-3.5 2024.01		68
ZS-Fusion 2024.01		66.7
DeBERTa-v3-L (CANDLE Distilled) 2024.01		65.9
CAR-ROBERTa-L 2024.01		64.8
Baseline 2025.05		64.8
STL-Adapter 2024.01		64.7
STL-Adapter 2024.01		64.4
CAR-ROBERTa-L 2024.01		64
CAR-DEBERTa-v3-L 2024.01		64
CAR-DEBERTa-v3-L 2024.01		63.8
STL-PLM 2024.01		63.2
ROBERTa-L (MR) 2024.01		63.2
ROBERTa-L (MR) 2024.01		63.1
DeBERTa-v3-L (MR) 2024.01		62.1
MTL 2024.01		61.9
ROBERTa-L (MR) 2024.01		61
DeBERTa-v3-L (MR) 2024.01		59.7
VERA-T5-xxl (CANDLE Distilled) 2024.01		59.4
Tensor RMS + C 2025.05		58.6
Task-specific FT 2026.02		58.39
VERA-T5-xxl 2024.01		58.2
VERA-T5-xxl 2024.01		58.1
TSV-M 2026.02		58.03
VERA-T5-xxl 2024.01		57.7
GPT-4 2024.01		57
TIES 2026.02		56.96
Task-specific FT 2026.02		56.81
MICO 2024.01		56
OrthoMerge 2026.02		55.17
TSV-M 2026.02		55.02
ROBERTa-L (MR) 2024.01		54.8
ProQA 2023.05		54.5
TA 2026.02		54.15
DARE 2026.02		53.84
TIES 2026.02		53.74
ZS-Fusion 2024.01		53.7
Tensor RMS + Sp 2025.05		53.6
Muppet 2023.05		52.63
OLTQA 2023.05		52.51
Hyperformer++ 2023.05		52.46
TA 2026.02		52.2
DARE 2026.02		52.1
MTL 2024.01		52
OLTQA 2023.05		51.76
OLTQA 2023.05		51.48
OLTQA 2023.05		51.23
OLTQA 2023.05		50.72
OLTQA 2023.05		50.67
OLTQA 2023.05		50.41
LLAMA2 2024.01		50.3
UnifiedQA 2023.05		50.15
COMET-DynGen 2024.01		50.1
OLTQA 2023.05		49.85
Distributed Lion-MaVo 2024.03		49.75
G-Lion 2024.03		49.64
OLTQA 2023.05		49.64
G-AdamW 2024.03		48.97
LLAMA2 2024.01		48.3
Llama-3.1-8B 2026.02		48.11
Distributed Lion-Avg 2024.03		48.06
DeBERTa-v3-L 2024.01		47.8
ROBERTa-L 2024.01		47.3
Self-talk 2024.01		46.2
LLaMA 7B (Baseline) 2024.03		45.96
EPR 2023.05		45.96
GPT2-L 2024.01		44.6
Mistral-v0.1 2024.01		42.9
Block Absmax 2025.05		37.1
Channel Absmax 2025.05		35.3
Majority Vote 2024.01		33.6
Random Vote 2024.01		33.3
Tensor RMS 2025.05		32.9
Tensor Absmax 2025.05		32.2