NLP Wiki

User Tools

Site Tools

Trace: • classification • sign_language_processing • software • educational_framework_edf • conditional_random_field • compositional_generalization
nlp:compositional_generalization

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
nlp:compositional_generalization [2022/01/14 22:19] – [Papers] jmflanignlp:compositional_generalization [2023/07/07 18:21] (current) – [Papers using Curriculum Learning] jmflanig
Line 1: Line 1:
 ====== Compositional Generalization ====== ====== Compositional Generalization ======
  
 +===== Overviews =====
 +  * [[https://arxiv.org/pdf/2302.01067.pdf|Lin et al 2023 - A Survey on Compositional Generalization in Applications]] Not an NLP paper, and not very comprehensive.  WARNING: Missing a bunch of NLP work.
  
 ===== Papers ===== ===== Papers =====
Line 6: Line 8:
   * [[https://arxiv.org/pdf/1807.04640.pdf|Chang et al 2018 - Automatically Composing Representation Transformations as a Means for Generalization]] Has a good section on what compositional generalization is.  Positive results for curriculum learning   * [[https://arxiv.org/pdf/1807.04640.pdf|Chang et al 2018 - Automatically Composing Representation Transformations as a Means for Generalization]] Has a good section on what compositional generalization is.  Positive results for curriculum learning
   * [[https://arxiv.org/pdf/1906.05381.pdf|Lake 2019 - Compositional generalization through metasequence-to-sequence learning]]   * [[https://arxiv.org/pdf/1906.05381.pdf|Lake 2019 - Compositional generalization through metasequence-to-sequence learning]]
 +  * M-PCFGSET dataset: [[https://eliabruni.github.io/publications/hupkes2019compositionality.pdf|Hupkes et al 2019 - The Compositionality of Neural Networks: Integrating Symbolism and Connectionism]]
   * COGS dataset: [[https://arxiv.org/pdf/2010.05465.pdf|Kim & Linzen 2020 - COGS: A Compositional Generalization Challenge Based on Semantic Interpretation]]   * COGS dataset: [[https://arxiv.org/pdf/2010.05465.pdf|Kim & Linzen 2020 - COGS: A Compositional Generalization Challenge Based on Semantic Interpretation]]
   * [[https://arxiv.org/pdf/1912.09713.pdf|2019 - Measuring Compositional Generalization: A Comprehensive Method on Realistic Data]] Introduces MCD data split method   * [[https://arxiv.org/pdf/1912.09713.pdf|2019 - Measuring Compositional Generalization: A Comprehensive Method on Realistic Data]] Introduces MCD data split method
Line 11: Line 14:
   * [[https://arxiv.org/pdf/2009.06040.pdf|Herzig & Berant 2020 - Span-based Semantic Parsing for Compositional Generalization]]   * [[https://arxiv.org/pdf/2009.06040.pdf|Herzig & Berant 2020 - Span-based Semantic Parsing for Compositional Generalization]]
   * [[https://arxiv.org/pdf/2010.12725.pdf|Shaw et al 2020 - Compositional Generalization and Natural Language Variation: Can a Semantic Parsing Approach Handle Both?]] [[https://github.com/google-research/language/tree/master/language/nqg|Code and data]] Introduces TMCD split method   * [[https://arxiv.org/pdf/2010.12725.pdf|Shaw et al 2020 - Compositional Generalization and Natural Language Variation: Can a Semantic Parsing Approach Handle Both?]] [[https://github.com/google-research/language/tree/master/language/nqg|Code and data]] Introduces TMCD split method
-  * [[https://arxiv.org/pdf/2108.12284.pdf|Csordás et al 2021 - The Devil is in the Detail: Simple Tricks Improve Systematic Generalization of Transformers]]+  * Improving Transformers for COG: 
 +    * **[[https://arxiv.org/pdf/2108.12284.pdf|Csordás et al 2021 - The Devil is in the Detail: Simple Tricks Improve Systematic Generalization of Transformers]]** 
 +    * [[https://arxiv.org/pdf/2108.04378.pdf|Ontañón et al 2021 - Making Transformers Solve Compositional Tasks]]
  
 ==== Papers using Curriculum Learning ==== ==== Papers using Curriculum Learning ====
   * [[https://arxiv.org/pdf/2006.10627.pdf|2020 - Compositional Generalization by Learning Analytical Expressions]]   * [[https://arxiv.org/pdf/2006.10627.pdf|2020 - Compositional Generalization by Learning Analytical Expressions]]
 +
 +==== Prompting and Language Models ====
 +  * [[https://arxiv.org/pdf/2209.15003.pdf|Drozdov 2022 - Compositional Semantic Parsing with Large Language Models]]
 +
 +==== COG in Machine Translation ====
 +  * [[https://arxiv.org/pdf/2210.06709.pdf|Yin et al 2022 - Categorizing Semantic Representations for Neural Machine Translation]]
  
 ==== Non-NLP Papers ==== ==== Non-NLP Papers ====
Line 25: Line 36:
     * [[https://yale-lily.github.io/spider|Spider dataset]] [[https://arxiv.org/pdf/1809.08887.pdf|paper]]     * [[https://yale-lily.github.io/spider|Spider dataset]] [[https://arxiv.org/pdf/1809.08887.pdf|paper]]
   * CFQ dataset [[https://arxiv.org/pdf/1912.09713.pdf|paper]]   * CFQ dataset [[https://arxiv.org/pdf/1912.09713.pdf|paper]]
 +  * PCFG dataset [[https://arxiv.org/pdf/1908.08351.pdf|paper]] (A string edit operation composition benchmark)
 +  * M-PCFGSET dataset [[https://eliabruni.github.io/publications/hupkes2019compositionality.pdf|paper]]
   * Question Answering   * Question Answering
     * ComQA: Compositional Question-Answering dataset [[https://arxiv.org/pdf/2101.06400.pdf|paper]]     * ComQA: Compositional Question-Answering dataset [[https://arxiv.org/pdf/2101.06400.pdf|paper]]
Line 30: Line 43:
     * Small-scale MT dataset from the [[https://arxiv.org/pdf/1711.00350.pdf|SCAN dataset paper]]     * Small-scale MT dataset from the [[https://arxiv.org/pdf/1711.00350.pdf|SCAN dataset paper]]
     * CoGnition dataset [[https://arxiv.org/pdf/2105.14802.pdf|paper]]      * CoGnition dataset [[https://arxiv.org/pdf/2105.14802.pdf|paper]] 
 +
 +===== People =====
 +  * [[https://scholar.google.com/citations?user=dnZ8udEAAAAJ&hl=en|Jacob Andreas]] [[https://twitter.com/jacobandreas|Tw]]
  
 ===== Related Pages ===== ===== Related Pages =====
nlp/compositional_generalization.1642198788.txt.gz · Last modified: 2023/06/15 07:36 (external edit)
Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki