Misinformation Detection on SLN (test)

94.22Micro F1

KALM

Updated 4d ago

Evaluation Results

Method
KALM 2022.10	94.22	94.18	-	-
KALM 2022.10	94.22	94.18	94.33	94.22
KGAP 2022.10	92.17	92.3	-	-
KGAP 2022.10	92.17	92.3	92.67	92.17
Task-specific baseline SOTA 2022.10	89.17	89.12	-	-
CompareNet 2022.10	89.17	89.12	89.82	89.17
Pretrained Language Model (BestLM) 2022.10	88.17	88.46	-	-
GreaseLM+ 2022.10	88.17	88.64	-	-
RoBERTa 2022.10	88.17	87.34	89.02	88.17
GreaseLM+ 2022.10	88.17	88.64	88.56	88.17
LongFormer 2022.10	88	86.29	88.84	87.44
DeBERTa 2022.10	86.89	88.46	89.43	86.89
BART 2022.10	86.06	86.12	86.13	86.06
GCN + Attn 2022.10	85.27	85.24	85.59	85.27
GAT + Attn 2022.10	84.72	84.62	85.65	84.72
KELM 2022.10	84.11	82.8	-	-
KELM 2022.10	84.11	82.8	85.23	84.11
Joshi et al. 2022.10	82.72	83.98	-	-
Joshi et al. 2022.10	82.72	83.98	84.95	83.37
KnowBERT 2022.10	78.67	79.8	-	-
KnowBERT-W+W 2022.10	78.67	79.8	79.36	78.67
Electra 2022.10	75.44	67.53	83.22	75.44
KnowBERT-Wordnet 2022.10	74.72	72.74	77.22	74.72
GreaseLM 2022.10	73.83	75.2	-	-
GreaseLM 2022.10	73.83	75.2	74.33	73.83
KnowBERT-Wikidata 2022.10	72.17	69.41	73.57	72.17
Rubin et al. 2022.10	-	-	88	82