Math Word Problem Solving On Math23K

Metrics

Accuracy (training-test)

Results

Performance results of various models on this benchmark

		Paper Title
ATHENA (roberta-base)	84.4	ATHENA: Mathematical Reasoning with Thought Expansion
GTS w/ Data Augmentation	-	-
Generate and Rank	85.4	Generate & Rank: A Multi-task Framework for Math Word Problems
Exp-Tree	86.2	An Expression Tree Decoding Strategy for Mathematical Equation Generation
REAL2: Memory-augmented Solver	85.2	-
Graph2Tree	77.4	Graph-to-Tree Learning for Solving Math Word Problems
Multi-view* (ours)	87.1	Multi-View Reasoning: Consistent Contrastive Learning for Math Word Problem
SAU-Solver	-	Semantically-Aligned Universal Tree-Structured Solver for Math Word Problems
LBF	-	Learning by Fixing: Solving Math Word Problems with Weak Supervision
ATHENA (roberta-large)	86.5	ATHENA: Mathematical Reasoning with Thought Expansion
Recall and Learn	82.3	Recall and Learn: A Memory-augmented Solver for Math Word Problems
Roberta-DeductReasoner	-	Learning to Reason Deductively: Math Word Problem Solving as Complex Relation Extraction
GROUP-ATT	69.5	Modeling Intra-Relation in Math Word Problems with Different Functional Multi-Head Attentions
MWP-BERT	84.7	MWP-BERT: Numeracy-Augmented Pre-training for Math Word Problem Solving
RoBERTaGen	-	MWPToolkit: An Open-Source Framework for Deep Learning-Based Math Word Problem Solvers
GPT-4 (Teaching-Inspired)	-	Teaching-Inspired Integrated Prompting Framework: A Novel Approach for Enhancing Reasoning in Large Language Models
T-RNN	66.9	-
GTS	-	A Goal-Driven Tree-Structured Neural Model for Math Word Problems
Hybrid model w/ SNI	-	Deep Neural Solver for Math Word Problems

0 of 19 row(s) selected.