NGT @ ACL 2020

13 papers

Top Research Directions

The 36 research directions with NGT @ ACL 2020 papers, ranked by paper count. Share is the fraction of the edition’s papers in the direction.

#DirectionPapersShare
1Overcoming Hypothesis Redundancy in Sequence Decoding753.8%
2Progressive Domain Adaptation via Multi-Stage Fine-Tuning323.1%
3Overcoming the Parallel Document Scarcity Bottleneck215.4%
4Balancing Latency and Quality in Simultaneous Translation17.7%
5Balancing Style Transformation and Content Preservation17.7%
6Bias Propagation in Reward-Guided Generative Alignment17.7%
7Compressing Large Foundation Models via Knowledge Distillation17.7%
8Decomposing Context into Multi-Level Attention Hierarchies17.7%
9Decoupling Generation and Selection in Sequence Models17.7%
10Deep Encoders and Shallow Decoders for Efficient Inference17.7%
11Disentangling Content and Style in Latent Representations17.7%
12Document-Level Coherence in Machine Translation17.7%
13Episodic Simulation for Few-Shot Meta-Learning17.7%
14Exposure Bias in Autoregressive Sequence Generation17.7%
15First-Order Approximations for Efficient Meta-Learning17.7%
16How Domain Similarity Dictates Adaptation Strategy17.7%
17Initializing Student Networks with Pretrained Teacher Weights17.7%
18Learning Highly Adaptable Parameter Initializations17.7%
19Managing Domain Mismatch in Neural Machine Translation17.7%
20Overcoming the Computational Cost of Transformer Architectures17.7%
21Preserving Model Capabilities Under Aggressive Compression17.7%
22Probabilistic Knowledge Transfer via Soft Labels17.7%
23Progressive Domain Adaptation through Intermediate Distributions17.7%
24Right to Left Reranking for Sequence Generation17.7%
25Robustness to Input Noise in Neural Machine Translation17.7%

Download all directions as CSV

Institutions

Institutions credited by NGT @ ACL 2020 papers: a paper credits every institution named in its affiliations, once. 16 institutions share 13 credited papers.

#InstitutionPapersShare
1University of Illinois Urbana-Champaign 323.1%
2Carnegie Mellon University 215.4%
3Johns Hopkins University 215.4%
4Apple (United States) 17.7%
5Charles University 17.7%
6Duolingo (United States) 17.7%
7Heriot-Watt University 17.7%
8Ludwig-Maximilians-Universität München 17.7%
9Meta (United States) 17.7%
10Microsoft (United States) 17.7%
11Northeastern University 17.7%
12Nvidia (United States) 17.7%
13Tokyo Metropolitan University 17.7%
14University of Amsterdam 17.7%
15University of Edinburgh 17.7%
16University of Washington 17.7%

Download all institutions as CSV

About this data

Papers. NGT @ ACL 2020 lists 13 papers on Lacuna, duplicate versions counted once; rejected and withdrawn submissions are left out.

Directions. Directions are Lacuna’s clustering of the concept elements extracted from each paper, so a paper can belong to several and direction counts add up to more than the paper count. A direction’s share is its papers divided by the edition’s papers.

Institutions. A paper credits each institution named in its affiliations, resolved to a ROR record, once. 13 of 13 papers (100.0%) credit at least one institution and are the base for institution shares; the rest have no resolved affiliation yet.

Data. Counts come from Lacuna’s data build of 2026-11-19 and change as papers are added. Tables as CSV: directions · institutions. Cite as: Lacuna, “NGT @ ACL 2020”, https://lacuna.tiptreesystems.com/venue/ngt-acl/2020.

Top Authors

AuthorPapers
Aizhan Imankulova1
Alexandra Birch1
Amr Sharaf1
Chenglong Wang1
Chi Hu1
Christof Monz1
Graham Neubig1
Hal Daumé III1
Hiroaki Hayashi1
Huda Khayrallah1
Ioannis Konstas1
Jacob Bremerman1
Jan Kautz1
Jindřich Libovický1
Kenneth Heafield1
Kenton Murray1
Kevin Duh1
Kevin Lin1
Mamoru Komachi1
Marine Carpuat1

Impactful Papers

Papers from NGT @ ACL 2020 that shaped research clusters.

Influential

Papers

Title
A Deep Reinforced Model for Zero-Shot Cross-Lingual Summarization with Bilingual Semantic Similarity Rewards
Distill, Adapt, Distill: Training Small, In-Domain Models for Neural Machine Translation
English-to-Japanese Diverse Translation by Combining Forward and Backward Outputs
Expand and Filter: CUNI and LMU Systems for the WNGT 2020 Duolingo Shared Task
Findings of the Fourth Workshop on Neural Generation and Translation
Generating Diverse Translations via Weighted Fine-tuning and Hypotheses Filtering for the Duolingo STAPLE Task
Learning to Generate Multiple Style Transfer Outputs for an Input Sentence
Meta-Learning for Few-Shot NMT Adaptation
Simultaneous Translation and Paraphrase for Language Education
The JHU Submission to the 2020 Duolingo Shared Task on Simultaneous Translation and Paraphrase for Language Education
The NiuTrans System for WNGT 2020 Efficiency Task
The Unreasonable Volatility of Neural Machine Translation Models
Training and Inference Methods for High-Coverage Neural Machine Translation