Commonsense Question Answering

Benchmarks

Dataset Name	SOTA Method	Metric
CSQA (test)	Human	Accuracy0.953	127	3d ago
CommonSenseQA		Accuracy88.9	81	3d ago
CSQA	SAC Single-task	Accuracy82.72	44	3d ago
CosmosQA		Accuracy94	36	3d ago
SocialIQA (SIQA) (val)	ChatGPT + Chain-of-thought	Accuracy70.7	24	3d ago
CommonsenseQA (CSQA) (val)	ChatGPT + Self-consistent chain-of-thought	Accuracy75.7	23	3d ago
CommonsenseQA v1.0 (dev)	Our Model	Accuracy79.3	22	3d ago
WinoGrande (WG) (val)	DeBERTa-v3-L (CANDLE Distilled)	Accuracy78.3	21	3d ago
Abductive NLI (aNLI) (val)	DeBERTa-v3-L (CANDLE Distilled)	Accuracy0.812	21	3d ago
CommonsenseQA blind v1.0 (test)	Our Model	Accuracy75.3	20	3d ago
CSQA	QTALE	PIQA84.06	18	3d ago
MCEval CSQA 8K (test)	Magn-Probe	Accuracy84.6	14	3d ago
Commonsense QA		Reusability Score50.97	12	3d ago
CSQA	UDPO	Accuracy85.1	12	3d ago
CSQA2 (test)	UL20B	Accuracy70.1	11	3d ago
CSQA (OOD)	R1 Distill -> GRPO	Accuracy63.8	10	3d ago
ARC-E	Dense	Accuracy72.31	8	3d ago
ARC-C-ZH	Dense	Score33.96	8	3d ago
ARC-C	PHSA	Accuracy41.13	8	3d ago
Scientific Commonsense (QASC) 1.0 (dev)	GPT-3	Accuracy55.18	8	3d ago
CommonsenseQA (CSQA2) 2.0 (dev)	ELABOR	Accuracy58.72	8	3d ago
ECQA (test)	Llama2-70B	Accuracy79.7	7	3d ago
QASC (dev)	CPACE	Accuracy83.7	7	3d ago
CSQA Synonym Replacement WordNet-based (test)	G-DAUG-Rand	Accuracy72.1	6	3d ago
OpenBookQA (OBQA) 1.0 (test)	GPT-3	Accuracy59.4	5	3d ago

Showing 25 of 34 rows