Croatia

This study presents an analytical evaluation of neural text simplification (TS) systems. Because recent TS models are trained in an end-to-end fashion, it is difficult to grasp their abilities to perform particular simplification operations. For the advancement of TS research and development, we should understand in detail what current TS systems can and cannot perform in comparison with human performance. To that end, we first developed an analytical evaluation framework consisting of fine-grained taxonomies of simplification strategies (at both the surface and content levels) and errors. Using this framework, we annotated TS instances produced by professional human editors and multiple neural TS systems and compared the results. Our analyses concretely and quantitatively revealed a wide gap between humans and systems, specifically indicating that systems tend to perform deletions and local substitutions while excessively omitting important information, and that the systems can hardly perform information addition operations. Based on our analyses, we also provide detailed directions to address these limitations.

EACL 2023

Gauging the Gap Between Human and Machine Text Simplification Through Analytical Evaluation of Simplification Strategies and Errors

simplification strategy

analytical evaluation

text simplification

findings / work in progress

### Welcome to the 17th Conference of the European Chapter of the Association for Computational Linguistics (EACL) 
Continuing its mission of expanding and involving the science community of all European countries, EACL had selected the Ukrainian community for the 16th EACL, which took place online due to the COVID pandemic. Unfortunately, the ongoing war made the organisation in Kyiv impossible. Considering the importance of physical interaction among researchers, especially after the restrictions imposed by the COVID pandemic, in addition to an online mode, the conference will be held in Dubrovnik, Croatia, from 2 to 6 of May, 2023. The original aim of strengthening the connection with the Ukrainian community will not change as our program will feature a dedicated session and a workshop to highlight work on Ukrainian language technologies. As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language.

You need to register for the conference in order to access this site. Please visit https://2023.eacl.org/registration for more information.

As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language.

Commonsense generation aims to generate a plausible sentence containing all given unordered concept words. Previous methods focusing on this task usually directly concatenate these words as the input of a pre-trained language model (PLM). However, in PLMs' pre-training process, the inputs are often corrupted sentences with correct word order. This input distribution discrepancy between pre-training and fine-tuning makes the model difficult to fully utilize the knowledge of PLMs. In this paper, we propose a two-stage framework to alleviate this issue. Firstly, in pre-training stage, we design a new format of input to endow PLMs the ability to deal with masked sentences with incorrect word order. Secondly, during fine-tuning, we insert the special token [MASK] between two consecutive concept words to make the input distribution more similar to the input distribution in pre-training. We conduct extensive experiments and provide thorough analysis to demonstrate the effectiveness of our proposed method.

Bridging the Gap between Pre-Training and Fine-Tuning for Commonsense Generation

Lifelogging has gained more attention due to its wide applications, such as personalized recommendations or memory assistance. The issues of collecting and extracting personal life events have emerged. People often share their life experiences with others through conversations. However, extracting life events from conversations is rarely explored. In this paper, we present Life Event Dialog, a dataset containing fine-grained life event annotations on conversational data. In addition, we initiate a novel Conversational Life Event Extraction task and differentiate the task from the public event extraction or the life event extraction from other sources like microblogs. We explore three information extraction (IE) frameworks to address the Conversational Life Event Extraction task: OpenIE, relation extraction, and event extraction. A comprehensive empirical analysis of the three baselines is established. The results suggest that the current event extraction model still struggles with extracting life events from human daily conversations. Our proposed Life Event Dialog dataset and in-depth analysis of IE frameworks will facilitate future research on life event extraction from conversations.

LED: A Dataset for Life Event Extraction from Dialogs

Evidence data for automated fact-checking (AFC) can be in multiple modalities such as text, tables, images, audio, or video. While there is increasing interest in using images for AFC, previous works mostly focus on detecting manipulated or fake images. 
We propose a novel task, chart-based fact-checking, and introduce ChartBERT as the first model for AFC against chart evidence. ChartBERT leverages textual, structural and visual information of charts to determine the veracity of textual claims. For evaluation, we create ChartFC, a new dataset of 15,886 charts. We systematically evaluate 75 different vision-language (VL) baselines and show that ChartBERT outperforms VL models, achieving 63.8% accuracy. 
Our results suggest that the task is complex yet feasible, with many challenges ahead.

Reading and Reasoning over Chart Images for Evidence-based Automated Fact-Checking

Reinforcement learning has been widely adopted to model dialogue managers in task-oriented dialogues. However, the user simulator provided by state-of-the-art dialogue frameworks are only rough approximations of human behaviour. The ability to learn from a small number of human interactions is hence crucial, especially on multi-domain and multi-task environments where the action space is large. We therefore propose to use structured policies to improve sample efficiency when learning on these kinds of environments. We also evaluate the impact of learning from human vs simulated experts. Among the different levels of structure that we tested, the graph neural networks (GNNs) show a remarkable superiority by reaching a success rate above 80% with only 50 dialogues when learning from simulated experts. They also show superiority when learning from human experts, although a performance drop was observed. We therefore suggest to concentrate future research efforts on bridging the gap between human data, simulators and automatic evaluators in dialogue frameworks.

Few-Shot Structured Policy Learning for Multi-Domain and Multi-Task Dialogues

Recent advancements in Large Language Models (LLMs) have drawn increasing attention since the learned embeddings pretrained on large-scale datasets have shown powerful ability in various downstream applications. However, whether the learned knowledge by LLMs can be transferred to clinical cardiology remains unknown. In this work, we aim to bridge this gap by transferring the knowledge of LLMs to clinical Electrocardiography (ECG). We propose an approach for cardiovascular disease diagnosis and automatic ECG diagnosis report generation. We also introduce an additional loss function by Optimal Transport (OT) to align the distribution between ECG and language embedding. 
The learned embeddings are evaluated on two downstream tasks: (1) automatic ECG diagnosis report generation, and (2) zero-shot cardiovascular disease detection. Our approach is able to generate high-quality cardiac diagnosis reports and also achieves competitive zero-shot classification performance even compared with supervised baselines, which proves the feasibility of transferring knowledge from LLMs to the cardiac domain.

Transfer Knowledge from Natural Language to Electrocardiography: Can We Detect Cardiovascular Disease Through Language Models?

Real-world applications of language models entail data privacy constraints when learning from diverse data domains. Federated learning with pretrained language models for language tasks has been gaining attention lately but there are definite confounders that warrants a careful study. Specifically, understanding the limits of federated NLP applications through varying the effects of different aspects (such as data heterogeneity, the trade-off between training time and performance, the effect of different data, and client distributions and sensitivity of the shared model to learning local distributions) is necessary to evaluate whether language models indeed learn to generalize by adapting to the different domains. Towards that, we elaborate different hypotheses over the components in federated NLP architectures and study them in detail with relevant experiments over three tasks: Stanford Sentiment Treebank-2, OntoNotes-5.0 and GigaWord. The experiments with different Transformer inductive biases on the variety of tasks provide a glimpse at the understanding of federated learning at NLP tasks. Specifically, the analysis suggests that regularization due to the ensembling effect may be masquerading as domain adaptation of federated learning in NLP with pre-trained language models.

Practical Takes on Federated Learning with Pretrained Language Models

Propaganda aims to persuade an audience by appealing to emotions and using faulty reasoning, with the purpose of promoting a particular point of view. Similarly, metaphor modifies the semantic frame, thus eliciting a response that can be used to tune up or down the emotional volume of the message. Given the close relationship between them, we hypothesize that, when modeling them computationally, it can be beneficial to do so jointly. In particular, we perform multi-task learning with propaganda identification as the main task and metaphor detection as an auxiliary task. To the best of our knowledge, this is the first work that models metaphor and propaganda together. We experiment with two datasets for identifying propaganda techniques in news articles and in memes shared on social media. We find that leveraging metaphor improves model performance, particularly for the two most common propaganda techniques: loaded language and name-calling.

Paper Bullets: Modeling Propaganda with the Help of Metaphor

Large neural language models (LLMs) can be powerful tools for research in lexical semantics. We illustrate this potential using the English verb "break", which has numerous senses and appears in a wide range of syntactic frames. We show that LLMs capture known sense distinctions and can be used to identify informative new sense combinations for further analysis. More generally, we argue that LLMs are aligned with lexical semantic theories in providing high-dimensional, contextually modulated representations, but LLMs' lack of discrete features and dependence on usage-based data offer a genuinely new perspective on traditional problems in lexical semantics.

Lexical Semantics with Large Language Models: A Case Study of English "break"

Missing information is a common issue of dialogue summarization where some information in the reference summaries is not covered in the generated summaries. To address this issue, we propose to utilize natural language inference (NLI) models to improve coverage while avoiding introducing factual inconsistencies. Specifically, we use NLI to compute fine-grained training signals to encourage the model to generate content in the reference summaries that have not been covered, as well as to distinguish between factually consistent and inconsistent generated sentences. Experiments on the DialogSum and SAMSum datasets confirm the effectiveness of the proposed approach in balancing coverage and faithfulness, validated with automatic metrics and human evaluations. Additionally, we compute the correlation between commonly used automatic metrics with human judgments in terms of three different dimensions regarding coverage and factual consistency to provide insight into the most suitable metric for evaluating dialogue summaries.

SWING: Balancing Coverage and Faithfulness for Dialogue Summarization

Cloze tests play an essential role in language assessment and help language learners improve their skills.
In this paper, we propose a novel task called Cloze Quality Estimation (CQE) --- a zero-shot task of evaluating whether a cloze test is of sufficient "high-quality'' for language assessment based on two important factors: reliability and validity.
We have taken the first step by creating a new dataset named CELA for the CQE task, which includes English cloze tests and corresponding evaluations about their quality annotated by native English speakers, which includes 2,597 and 1,730 instances in aspects of reliability and validity, respectively.
We have tested baseline evaluation methods on the dataset, showing that our method could contribute to the CQE task, but the task is still challenging.

Premium content

Gauging the Gap Between Human and Machine Text Simplification Through Analytical Evaluation of Simplification Strategies and Errors

Downloads

Next from EACL 2023

Bridging the Gap between Pre-Training and Fine-Tuning for Commonsense Generation

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES