AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Two Birds with One Stone: Investigating Invertible Neural Networks for Inverse Problems in Morphology

Gözde Gül Şahin, Iryna Gurevych

AAAI 2020 · Architectures and efficiency › RNNs and normalization · approach: Brain-inspired recurrent architectures · 4 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
Use one INN trained in both directions with Ly + LxMorphological inflection · Finnish, French, Irish, Romanian, Tagalog, Turkishsurface exact match (%): 85.15 -> 92.23; 88.94 -> 97.55; 61.11 -> 79.63; 89.45 -> 97.67; 30.05 -> 33.33; 88.89 -> 97.45Improved
Add reverse loss Lx to forward-only training (ablation)INN morphological inflection · Finnishlemma exact match (%): 0.01 -> 95.56Improved
Add reverse loss Lx to forward-only training (ablation)INN morphological inflection · Frenchlemma exact match (%): 0.0 -> 98.56Improved
Add reverse loss Lx to forward-only training (ablation)INN morphological inflection · Irishlemma exact match (%): 0.0 -> 97.22Improved
Add reverse loss Lx to forward-only training (ablation)INN morphological inflection · Romanianlemma exact match (%): 0.0 -> 99.26Improved
Add reverse loss Lx to forward-only training (ablation)INN morphological inflection · Tagaloglemma exact match (%): 0.0 -> 85.41Improved
Add reverse loss Lx to forward-only training (ablation)INN morphological inflection · Turkishlemma exact match (%): 0.0 -> 99.62Improved
Add categorical latent variables with Lz (ablation)INN lemmatization · All six languageslemma exact match (%): The paper reports improvements in all languages; see Table 2Improved
Increase latent dimension and category count from z dim 2/categories 3 to z dim 6/categories 4 (ablation)INN lemmatization · Finnish, French, Irish, Romanian, Tagalog, Turkishlemma exact match (%): 97.54 -> 98.18; 96.03 -> 96.14; 94.2 -> 95.65; 88.85 -> 90.95; 90.03 -> 91.52; 99.01 -> 99.54Improved
Add tag loss Lt (ablation)INN morphological inflection with Ly + Lx · Finnish and Tagaloglemma exact match (%): 95.56 -> 93.33; 85.41 -> 69.73Hurt
Add tag loss Lt (ablation)INN morphological inflection with Ly + Lx · Irish and Turkishlemma exact match (%): 97.22 -> 100; 99.62 -> 99.74Improved
Add tag loss Lt (ablation)INN morphological inflection with Ly + Lx · Finnish, French, Irish, Romanian, Tagalog, Turkishtag F1 (%): 11.84 -> 12.42 -> 49.99; 23.5 -> 13.32 -> 26.01; 25.98 -> 20.55 -> 67.33; 16.5 -> 14.24 -> 0.02; 15.24 -> 10.13 -> 43.14Mixed

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