AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Joint Semantic Analysis with Document-Level Cross-Task Coherence Rewards

Rahul Aralikatte, Mostafa Abdou, Heather C Lent, Daniel Hershcovich, Anders Søgaard

AAAI 2021 · Vision-language and multimodal › Human and embodied modeling · approach: Layout-aware document understanding · 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
Fine-tune with document-level coherence rewardsCoreference resolver with LSTM + CNN · Six coreference datasetsmean F1: 49.01 -> 49.40Improved
Fine-tune with document-level coherence rewardsCoreference resolver with BERT-Tiny · Six coreference datasetsmean F1: 49.70 -> 50.95Improved
Fine-tune with document-level coherence rewardsCoreference resolver with BERT-Mini · Six coreference datasetsmean F1: 52.61 -> 52.88Improved
Fine-tune with document-level coherence rewardsCoreference resolver with BERT-Small · Six coreference datasetsmean F1: 52.76 -> 53.90Improved
Fine-tune with document-level coherence rewardsCoreference resolver with BERT-Medium · Six coreference datasetsmean F1: 55.67 -> 56.19Improved
Fine-tune with document-level coherence rewardsCoreference resolver with BERT-Base · Six coreference datasetsmean F1: 57.78 -> 58.18Improved
Fine-tune with document-level coherence rewardsSRL model with LSTM + CNN · Four SRL datasetsmean F1: 67.63 -> 67.74Improved
Fine-tune with document-level coherence rewardsSRL model with BERT-Tiny · Four SRL datasetsmean F1: 56.87 -> 57.08Improved
Fine-tune with document-level coherence rewardsSRL model with BERT-Mini · Four SRL datasetsmean F1: 70.51 -> 70.48Hurt
Fine-tune with document-level coherence rewardsSRL model with BERT-Small · Four SRL datasetsmean F1: 74.26 -> 74.48Improved
Fine-tune with document-level coherence rewardsSRL model with BERT-Medium · Four SRL datasetsmean F1: 75.62 -> 75.57Hurt
Fine-tune with document-level coherence rewardsSRL model with BERT-Base · Four SRL datasetsmean F1: 79.46 -> 79.52Improved

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