mlsanityData to 7 Oct 2026

Mimic and Conquer: Heterogeneous Tree Structure Distillation for Syntactic NLP

arXiv 2009.07411 ↗First posted 16 Sep 2020cs.CLcited by 15

Not on the map. Why?

Abstract (arXiv)

Syntax has been shown useful for various NLP tasks, while existing work mostly encodes singleton syntactic tree using one hierarchical neural network. In this paper, we investigate a simple and effective method, Knowledge Distillation, to integrate heterogeneous structure knowledge into a unified sequential LSTM encoder. Experimental results on four typical syntax-dependent tasks show that our method outperforms tree encoders by effectively integrating rich heterogeneous structure syntax, meanwhile reducing error propagation, and also outperforms ensemble methods, in terms of both the efficiency and accuracy.

Key facts. Cited by 15 papers in the map (data through 7 Oct 2026). First posted on arXiv 16 Sep 2020.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2022: 3; 2023: 12.

15 papers.

Sort:

  1. 2023Scene Graph as Pivoting: Inference-time Image-free Unsupervised Multimodal Machine Translation with Visual Scene Hallucinationcited by 27
  2. 2023On the Robustness of Aspect-based Sentiment Analysis: Rethinking Model, Data, and Trainingcited by 20
  3. 2022Entity-centered Cross-document Relation Extractioncited by 18
  4. 2023Cross2StrA: Unpaired Cross-lingual Image Captioning with Cross-lingual Cross-modal Structure-pivoted Alignmentcited by 15
  5. 2023Constructing Holistic Spatio-Temporal Scene Graph for Video Semantic Role Labelingcited by 15
  6. 2022Conversational Semantic Role Labeling with Predicate-Oriented Latent Graphcited by 14
  7. 2023Revisiting Disentanglement and Fusion on Modality and Context in Conversational Multimodal Emotion Recognitioncited by 13
  8. 2023Constructing Code-mixed Universal Dependency Forest for Unbiased Cross-lingual Relation Extractioncited by 6
  9. 2023ECQED: Emotion-Cause Quadruple Extraction in Dialogscited by 2
  10. 2023TKDP: Threefold Knowledge-enriched Deep Prompt Tuning for Few-shot Named Entity Recognitioncited by 2
  11. 2023Revisiting Conversation Discourse for Dialogue Disentanglementcited by 2
  12. 2023A Bi-directional Multi-hop Inference Model for Joint Dialog Sentiment Classification and Act Recognitioncited by 2
3 more
  1. 2023Can We Use Probing to Better Understand Fine-tuning and Knowledge Distillation of the BERT NLU?cited by 1
  2. 2023DialogRE^C+: An Extension of DialogRE to Investigate How Much Coreference Helps Relation Extraction in Dialogscited by 1
  3. 2022ExpertNet: A Symbiosis of Classification and Clusteringcited by 0