Papers by Zichao Yang

10 papers
Composable Text Controls in Latent Space with ODEs (2023.emnlp-main)

Copied to clipboard

Challenge: Existing approaches to composable text operations often require plug-and-play . a single LM can perform arbitrary text operation composition in the latent space .
Approach: They propose an efficient approach for composable text operations in the latent space of text . they connect pretrained LMs to the laten space and adapt them to the space .
Outcome: The proposed approach improves on existing methods in the latent space of text.
Texar: A Modularized, Versatile, and Extensible Toolkit for Text Generation (P19-3)

Copied to clipboard

Challenge: Texar is an open-source text generation toolkit that supports a broad set of text generation tasks.
Approach: They introduce Texar, an open-source text generation toolkit that supports text generation tasks.
Outcome: Texar supports machine translation, summarization, dialog, content manipulation, and more.
Interpretable Math Word Problem Solution Generation via Step-by-step Planning (2023.acl-long)

Copied to clipboard

Challenge: Existing approaches to solving math word problems focus on obtaining the correct answer.
Approach: They propose a step-by-step planning approach for intermediate solution generation that strategically plans the generation of the next solution step based on the MWP and the previous solution steps.
Outcome: The proposed approach improves the accuracy and interpretability of the solution on automatic metrics and human evaluation.
MixText: Linguistically-Informed Interpolation of Hidden Space for Semi-Supervised Text Classification (2020.acl-main)

Copied to clipboard

Challenge: Xie et al., 2016) demonstrate that semi-supervised learning models suffer from over-fitting when there is only limited labeled data.
Approach: They propose a semi-supervised learning method for text classification using a data augmentation method called TMix.
Outcome: The proposed method outperforms pre-trained and fine-tuned models on several text classification benchmarks.
Local Additivity Based Data Augmentation for Semi-supervised NER (2020.emnlp-main)

Copied to clipboard

Challenge: Named Entity Recognition (NER) is one of the first stages in deep language understanding yet current NER models heavily rely on human-annotated data.
Approach: They propose a Local Additivity based Data Augmentation method for semi-supervised Named Entity Recognition (NER) that creates virtual samples by interpolating sequences close to each other.
Outcome: The proposed method improves both entity and context learning by adding to training data and extending it to semi-supervised setting.
Don’t Take It Literally: An Edit-Invariant Sequence Loss for Text Generation (2022.naacl-main)

Copied to clipboard

Challenge: Neural text generation models are typically trained by maximizing log-likelihood with the sequence cross entropy (CE) loss.
Approach: They propose an Edit-Invariant Sequence Loss method which computes the matching loss of a target sequence with all n-grams in the generated sequence.
Outcome: The proposed method outperforms the common CE loss and strong baselines on a wide range of tasks.
TreeMix: Compositional Constituency-based Data Augmentation for Natural Language Understanding (2022.naacl-main)

Copied to clipboard

Challenge: Existing data augmentation methods miss the important characteristic of compositionality, meaning of a complex expression is built from its sub-parts.
Approach: They propose a compositional data augmentation approach for natural language understanding called TreeMix that leverages constituency parsing tree to decompose sentences into constituent sub-structures and the Mixup data enhancing technique to recombine them to generate new sentences.
Outcome: The proposed approach outperforms current state-of-the-art methods on text classification and SCAN.
Data-to-Text Generation with Style Imitation (2020.findings-emnlp)

Copied to clipboard

Challenge: Recent approaches to data-to-text generation focus on improving content fidelity, but lack explicit control over writing styles.
Approach: They propose a way to control writing styles by using existing sentences as "soft" templates . they conduct experiments in restaurants and sports domains to test their approach .
Outcome: The proposed approach achieves stronger performance than a range of comparison methods.
Progressive Generation of Long Text with Pretrained Language Models (2021.naacl-main)

Copied to clipboard

Challenge: Existing methods for "long" text generation are limited to outputs of 50-200 tokens . however, our proposed ProGen generates coherent long passages of text in a progressive manner .
Approach: They propose a method for generating coherent long passages of text in a progressive manner . they first produce domain-specific content keywords and then refine them into complete passages . human evaluation validates that their proposed generation is more coherent .
Outcome: The proposed method produces domain-specific content keywords and refines them into complete passages in multiple stages.
Let’s Make Your Request More Persuasive: Modeling Persuasive Strategies via Semi-Supervised Neural Nets on Crowdfunding Platforms (N19-1)

Copied to clipboard

Challenge: Existing models can't quantify persuasiveness of requests or extract successful persuasive strategies.
Approach: They propose a semi-supervised hierarchical neural network model to quantify persuasiveness and identify persuasive strategies in advocacy requests.
Outcome: The proposed method outperforms baseline models and offers increased interpretability of persuasive speech.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations