Automatic methods for generating and gathering linguistic data have proven effective for fine-tuning Language Models (LMs) in languages less resourced than English. Still, while there has been emphasis on data quantity, less attention has been given to its quality. In this work, we investigate the impact of human intervention on machine-generated data when fine-tuning dialogical models. In particular, we study (1) whether post-edited dialogues exhibit higher perceived quality compared to the originals that were automatically generated; (2) whether fine-tuning with post-edited dialogues results in noticeable differences in the generated outputs; and (3) whether post-edited dialogues influence the outcomes when considering the parameter size of the LMs. To this end we created HED-IT, a large-scale dataset where machine-generated dialogues are paired with the version post-edited by humans. Using both the edited and unedited portions of HED-IT, we fine-tuned three different sizes of an LM. Results from both human and automatic evaluation show that the different quality of training data is clearly perceived and it has an impact also on the models trained on such data. Additionally, our findings indicate that larger models are less sensitive to data quality, whereas this has a crucial impact on smaller models. These results enhance our comprehension of the impact of human intervention on training data in the development of high-quality LMs.

Large language models (LLMs) with Chain-of-thought (CoT) have recently emerged as a powerful technique for eliciting reasoning to improve various downstream tasks. As most research mainly focuses on English, with few explorations in a multilingual context, the question of how reliable this reasoning capability is in different languages is still open. To address it directly, we study multilingual reasoning consistency across multiple languages, using popular open-source LLMs. First, we compile the first large-scale multilingual math reasoning dataset, *mCoT-MATH*, covering eleven diverse languages. Then, we introduce multilingual CoT instruction tuning to boost reasoning capability across languages, thereby improving model consistency. While existing LLMs show substantial variation across the languages we consider, and especially low performance for lesser resourced languages, our 7B parameter model *mCoT* achieves impressive consistency across languages, and superior or comparable performance to close- and open-source models even of much larger sizes.

In the context of the ReproHum project aimed at assessing the reliability of human evaluation, we replicated the human evaluation conducted in “Generating Scientific Definitions with Controllable Complexity” by August et al. (2022). Specifically, humans were asked to assess the fluency of automatically generated scientific definitions by three different models, with output complexity varying according to target audience. Evaluation conditions were kept as close as possible to the original study, except of necessary and minor adjustments. Our results, despite yielding lower absolute performance, show that relative performance across the three tested systems remains comparable to what was observed in the original paper. On the basis of lower inter-annotator agreement and feedback received from annotators in our experiment, we also observe that the ambiguity of the concept being evaluated may play a substantial role in human assessment.

In spite of the core role human judgement plays in evaluating the performance of NLP systems, the way human assessments are elicited in NLP experiments, and to some extent the nature of human judgement itself, pose challenges to the reliability and validity of human evaluation. In the context of the larger ReproHum project, aimed at running large scale multi-lab reproductions of human judgement, we replicated the understandability assessment by humans on several generated outputs of simplified text described in the paper “Neural Text Simplification of Clinical Letters with a Domain Specific Phrase Table” by Shardlow and Nawaz, appeared in the Proceedings of ACL 2019. Although we had to implement a series of modifications compared to the original study, which were necessary to run our human evaluation on exactly the same data, we managed to collect assessments and compare results with the original study. We obtained results consistent with those of the reference study, confirming their findings. The paper is complete with as much information as possible to foster and facilitate future reproduction.

We report our efforts in identifying a set of previous human evaluations in NLP that would be suitable for a coordinated study examining what makes human evaluations in NLP more/less reproducible. We present our results and findings, which include that just 13% of papers had (i) sufficiently low barriers to reproduction, and (ii) enough obtainable information, to be considered for reproduction, and that all but one of the experiments we selected for reproduction was discovered to have flaws that made the meaningfulness of conducting a reproduction questionable. As a result, we had to change our coordinated study design from a reproduce approach to a standardise-then-reproduce-twice approach. Our overall (negative) finding that the great majority of human evaluations in NLP is not repeatable and/or not reproducible and/or too flawed to justify reproduction, paints a dire picture, but presents an opportunity for a rethink about how to design and report human evaluations in NLP.

We reproduced the human-based evaluation of the continuation of narratives task presented by Chakrabarty et al. (2022). This experiment is performed as part of the ReproNLP Shared Task on Reproducibility of Evaluations in NLP (Track C). Our main goal is to reproduce the original study under conditions as similar as possible. Specifically, we follow the original experimental design and perform human evaluations of the data from the original study, while describing the differences between the two studies. We then present the results of these two studies together with an analysis of similarities between them. Inter-annotator agreement (Krippendorff’s alpha) in the reproduction study is lower than in the original study, while the human evaluation results of both studies have the same trends, that is, our results support the findings in the original study.

Pre-trained language models (PLMs) have achieved great success in NLP and have recently been used for tasks in computational semantics. However, these tasks do not fully benefit from PLMs since meaning representations are not explicitly included. We introduce multilingual pre-trained language-meaning models based on Discourse Representation Structures (DRSs), including meaning representations besides natural language texts in the same model, and design a new strategy to reduce the gap between the pre-training and fine-tuning objectives. Since DRSs are language neutral, cross-lingual transfer learning is adopted to further improve the performance of non-English tasks. Automatic evaluation results show that our approach achieves the best performance on both the multilingual DRS parsing and DRS-to-text generation tasks. Correlation analysis between automatic metrics and human judgements on the generation task further validates the effectiveness of our model. Human inspection reveals that out-of-vocabulary tokens are the main cause of erroneous results.

Different ways of linguistically expressing the same real-world event can lead to different perceptions of what happened. Previous work has shown that different descriptions of gender-based violence (GBV) influence the reader’s perception of who is to blame for the violence, possibly reinforcing stereotypes which see the victim as partly responsible, too. As a contribution to raise awareness on perspective-based writing, and to facilitate access to alternative perspectives, we introduce the novel task of automatically rewriting GBV descriptions as a means to alter the perceived level of blame on the perpetrator. We present a quasi-parallel dataset of sentences with low and high perceived responsibility levels for the perpetrator, and experiment with unsupervised (mBART-based), zero-shot and few-shot (GPT3-based) methods for rewriting sentences. We evaluate our models using a questionnaire study and a suite of automatic metrics.

Figures of speech help people express abstract concepts and evoke stronger emotions than literal expressions, thereby making texts more creative and engaging. Due to its pervasive and fundamental character, figurative language understanding has been addressed in Natural Language Processing, but it’s highly understudied in a multilingual setting and when considering more than one figure of speech at the same time. To bridge this gap, we introduce multilingual multi-figurative language modelling, and provide a benchmark for sentence-level figurative language detection, covering three common figures of speech and seven languages. Specifically, we develop a framework for figurative language detection based on template-based prompt learning. In so doing, we unify multiple detection tasks that are interrelated across multiple figures of speech and languages, without requiring task- or language-specific modules. Experimental results show that our framework outperforms several strong baselines and may serve as a blueprint for the joint modelling of other interrelated tasks.

Although text style transfer has witnessed rapid development in recent years, there is as yet no established standard for evaluation, which is performed using several automatic metrics, lacking the possibility of always resorting to human judgement. We focus on the task of formality transfer, and on the three aspects that are usually evaluated: style strength, content preservation, and fluency. To cast light on how such aspects are assessed by common and new metrics, we run a human-based evaluation and perform a rich correlation analysis. We are then able to offer some recommendations on the use of such metrics in formality transfer, also with an eye to their generalisability (or not) to related tasks.

Figurative language generation is the task of reformulating a given text in the desired figure of speech while still being faithful to the original context. We take the first step towards multi-figurative language modelling by providing a benchmark for the automatic generation of five common figurative forms in English. We train mFLAG employing a scheme for multi-figurative language pre-training on top of BART, and a mechanism for injecting the target figurative information into the encoder; this enables the generation of text with the target figurative form from another figurative form without parallel figurative-figurative sentence pairs. Our approach outperforms all strong baselines. We also offer some qualitative analysis and reflections on the relationship between the different figures of speech.

We exploit the pre-trained seq2seq model mBART for multilingual text style transfer. Using machine translated data as well as gold aligned English sentences yields state-of-the-art results in the three target languages we consider. Besides, in view of the general scarcity of parallel data, we propose a modular approach for multilingual formality transfer, which consists of two training strategies that target adaptation to both language and task. Our approach achieves competitive performance without monolingual task-specific parallel data and can be applied to other style transfer tasks as well as to other languages.

We ask subjects whether they perceive as human-produced a bunch of texts, some of which are actually human-written, while others are automatically generated. We use this data to fine-tune a GPT-2 model to push it to generate more human-like texts, and observe that this fine-tuned model produces texts that are indeed perceived more human-like than the original model. Contextually, we show that our automatic evaluation strategy well correlates with human judgements. We also run a linguistic analysis to unveil the characteristics of human- vs machine-perceived language.

Style transfer aims to rewrite a source text in a different target style while preserving its content. We propose a novel approach to this task that leverages generic resources, and without using any task-specific parallel (source–target) data outperforms existing unsupervised approaches on the two most popular style transfer tasks: formality transfer and polarity swap. In practice, we adopt a multi-step procedure which builds on a generic pre-trained sequence-to-sequence model (BART). First, we strengthen the model’s ability to rewrite by further pre-training BART on both an existing collection of generic paraphrases, as well as on synthetic pairs created using a general-purpose lexical resource. Second, through an iterative back-translation approach, we train two models, each in a transfer direction, so that they can provide each other with synthetically generated pairs, dynamically in the training process. Lastly, we let our best resulting model generate static synthetic pairs to be used in a supervised training regime. Besides methodology and state-of-the-art results, a core contribution of this work is a reflection on the nature of the two tasks we address, and how their differences are highlighted by their response to our approach.

Scarcity of parallel data causes formality style transfer models to have scarce success in preserving content. We show that fine-tuning pre-trained language (GPT-2) and sequence-to-sequence (BART) models boosts content preservation, and that this is possible even with limited amounts of parallel data. Augmenting these models with rewards that target style and content –the two core aspects of the task– we achieve a new state-of-the-art.

An ongoing debate in the NLG community concerns the best way to evaluate systems, with human evaluation often being considered the most reliable method, compared to corpus-based metrics. However, tasks involving subtle textual differences, such as style transfer, tend to be hard for humans to perform. In this paper, we propose an evaluation method for this task based on purposely-trained classifiers, showing that it better reflects system differences than traditional metrics such as BLEU.