Roll call vote prediction (Time-Based)

94.46Balanced Accuracy

KALM

Updated 3mo ago

Evaluation Results

Method
KALM 2022.10	94.46	91.97
KALM 2022.10	94.46	91.97
Joshi et al. 2022.10	92.63	89.31
Joshi et al. 2022.10	92.63	89.31
GreaseLM+ 2022.10	91.69	87.95
GreaseLM+ 2022.10	91.69	87.95
KELM 2022.10	90.8	86.62
KELM 2022.10	90.8	86.62
RoBERTa 2022.10	90.4	84.78
Pretrained Language Model (BestLM) 2022.10	90.4	85.21
BART 2022.10	90.25	85.21
PAR 2022.10	89.92	-
Task-specific baseline SOTA 2022.10	89.92	84.35
Vote 2022.10	89.76	84.35
LongFormer 2022.10	89.32	83.42
Electra 2022.10	88.92	82.5
DeBERTa 2022.10	88.59	81.38
GreaseLM 2022.10	88.21	79.73
GreaseLM 2022.10	88.21	79.73
KnowBERT-W+W 2022.10	87.07	78.42
KnowBERT 2022.10	87.07	78.9
KnowBERT-Wordnet 2022.10	86.92	78.9
KnowBERT-Wikidata 2022.10	86.45	78.21
ideal-vector 2022.10	81.95	75.49
KGAP 2022.10	77.9	70.81
KGAP 2022.10	77.9	70.81