AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Distributed Representations for Arithmetic Word Problems

Sowmya S Sundaram, Deepak P, Savitha Sam Abraham

AAAI 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Multilingual language representation pretraining · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add GCN layer to Bi-LSTM (ablation)Bi-LSTM Siamese · SingleOp schema retrievalP@3: 0.7272 -> 0.8114Improved
add Bi-LSTM to GCN (ablation)GCN Siamese · SingleOp schema retrievalP@3: 0.7503 -> 0.8114Improved
add GCN layer to Bi-LSTM (ablation)Bi-LSTM Siamese · SingleOp schema retrievalNDCG@10: 0.8618 -> 0.8876Improved
operator-based augmentation with Siamese model vs LSTM autoencoderlstm-ae · SingleOp schema retrievalP@10: 0.6413 -> 0.7129Improved
add Bi-LSTM to GCN (ablation)GCN Siamese · SingleOp classification AdaBoostaccuracy: 0.5664 -> 0.6726Improved
add GCN layer to Bi-LSTM (ablation)Bi-LSTM Siamese · SingleOp classification MLPaccuracy: 0.7434 -> 0.8761Improved
add Bi-LSTM to GCN (ablation)GCN Siamese · SingleOp classification Random Forestaccuracy: 0.8142 -> 0.8673Improved
use Ours instead of lstm-ae for random forest classificationlstm-ae · SingleOp classificationaccuracy: 0.8407 -> 0.8673Improved
use Ours instead of lstm-dae for equation-based retrieval P@10lstm-dae · SingleOp equation retrievalP@10: 0.4053 -> 0.5162Improved
use Ours instead of lstm-dae for equation-based retrieval NDCG@10lstm-dae · SingleOp equation retrievalNDCG@10: 0.7284 -> 0.8032Improved
transfer SingleOp-trained model to MAWPSOurs · MAWPS equation retrievalP@10: 0.2498 -> 0.3015Improved
transfer to MAWPSOurs · MAWPS equation retrievalP@3: 0.6797 -> 0.4862Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy