In this position paper, we argue that human evaluation of generative large language models (LLMs) should be a multidisciplinary undertaking that draws upon the insights from disciplines such as user experience research and human behavioral psychology to ensure that the experimental design and results are reliable. The conclusions from these evaluations, therefore, must consider factors such as usability, aesthetics and cognitive biases. We highlight how cognitive biases can conflate fluent information and truthfulness, and how cognitive uncertainty affects the reliability of rating scores such as Likert. Furthermore, the evaluation should differentiate the capabilities and weaknesses of increasingly powerful large language models - which requires effective test sets. Scalability of human evaluation is also crucial to wider adoption. Hence, to design an effective human evaluation system in the age of generative NLP we propose the ConSiDERS-The-Human evaluation framework consisting of 6 pillars - Consistency, Scoring Criteria, Differentiating, User Experience, Responsible, and Scalability.

Studies in bias and fairness in natural language processing have primarily examined social biases within a single language and/or across few attributes (e.g. gender, race). However, biases can manifest differently across various languages for individual attributes. As a result, it is critical to examine biases within each language and attribute. Of equal importance is to study how these biases compare across languages and how the biases are affected when training a model on multilingual data versus monolingual data. We present a bias analysis across Italian, Chinese, English, Hebrew, and Spanish on the downstream sentiment analysis task to observe whether specific demographics are viewed more positively. We study bias similarities and differences across these languages and investigate the impact of multilingual vs. monolingual training data. We adapt existing sentiment bias templates in English to Italian, Chinese, Hebrew, and Spanish for four attributes: race, religion, nationality, and gender. Our results reveal similarities in bias expression such as favoritism of groups that are dominant in each language’s culture (e.g. majority religions and nationalities). Additionally, we find an increased variation in predictions across protected groups, indicating bias amplification, after multilingual finetuning in comparison to multilingual pretraining.

Annotation errors that stem from various sources are usually unavoidable when performing large-scale annotation of linguistic data. In this paper, we evaluate the feasibility of using the Transformer model to detect various types of annotator errors in morphological data sets that contain inflected word forms. We evaluate our error detection model on four languages by introducing three different types of artificial errors in the data: (1) typographic errors, where single characters in the data are inserted, replaced, or deleted; (2) linguistic confusion errors where two inflected forms are systematically swapped; and (3) self-adversarial errors where the Transformer model itself is used to generate plausible-looking, but erroneous forms by retrieving high-scoring predictions from the search beam. Results show that the Transformer model can with perfect, or near-perfect recall detect errors in all three scenarios, even when significant amounts of the annotated data (5%-30%) are corrupted on all languages tested. Precision varies across the languages and types of errors, but is high enough that the model can be very effectively used to flag suspicious entries in large data sets for further scrutiny by human annotators.

Deep learning sequence models have been successful with morphological inflection generation. The SIGMORPHON shared task results in the past several years indicate that such models can perform well, but only if the training data covers a good amount of different lemmata, or if the lemmata to be inflected at test time have also been seen in training, as has indeed been largely the case in these tasks. Surprisingly, we find that standard models such as the Transformer almost completely fail at generalizing inflection patterns when trained on a limited number of lemmata and asked to inflect previously unseen lemmata—i.e. under “wug test”-like circumstances. This is true even though the actual number of training examples is very large. While established data augmentation techniques can be employed to alleviate this shortcoming by introducing a copying bias through hallucinating synthetic new word forms using the alphabet in the language at hand, our experiment results show that, to be more effective, the hallucination process needs to pay attention to substrings of syllable-like length rather than individual characters.

Part-of-Speech (POS) tags are routinely included as features in many NLP tasks. However, the importance and usefulness of POS tags needs to be examined as NLP expands to low-resource languages because linguists who provide many annotated resources do not place priority on early identification and tagging of POS. This paper describes an empirical study about the effect that POS tags have on two computational morphological tasks with the Transformer architecture. Each task is tested twice on identical data except for the presence/absence of POS tags, using published data in ten high- to low-resource languages or unpublished linguistic field data in five low-resource languages. We find that the presence or absence of POS tags does not have a significant bearing on performance. In joint segmentation and glossing, the largest average difference is an .09 improvement in F1-scores by removing POS tags. In reinflection, the greatest average difference is 1.2% in accuracy for published data and 5% for unpublished and noisy field data.

Backtranslation is a common technique for leveraging unlabeled data in low-resource scenarios in machine translation. The method is directly applicable to morphological inflection generation if unlabeled word forms are available. This paper evaluates the potential of backtranslation for morphological inflection using data from six languages with labeled data drawn from the SIGMORPHON shared task resource and unlabeled data from different sources. Our core finding is that backtranslation can offer modest improvements in low-resource scenarios, but only if the unlabeled data is very clean and has been filtered by the same annotation standards as the labeled data.

Analogy is assumed to be the cognitive mechanism speakers resort to in order to inflect an unknown form of a lexeme based on knowledge of other words in a language. In this process, an analogy is formed between word forms within an inflectional paradigm but also across paradigms. As neural network models for inflection are typically trained only on lemma-target form pairs, we propose three new ways to provide neural models with additional source forms to strengthen analogy-formation, and compare our methods to other approaches in the literature. We show that the proposed methods of providing a Transformer sequence-to-sequence model with additional analogy sources in the input are consistently effective, and improve upon recent state-of-the-art results on 46 languages, particularly in low-resource settings. We also propose a method to combine the analogy-motivated approach with data hallucination or augmentation. We find that the two approaches are complementary to each other and combining the two approaches is especially helpful when the training data is extremely limited.

This paper presents the submission by the CU Ling team from the University of Colorado to SIGMORPHON 2020 shared task 0 on morphological inflection. The task is to generate the target inflected word form given a lemma form and a target morphosyntactic description. Our system uses the Transformer architecture. Our overall approach is to treat the morphological inflection task as a paradigm cell filling problem and to design the system to leverage principal parts information for better morphological inflection when the training data is limited. We train one model for each language separately without external data. The overall average performance of our submission ranks the first in both average accuracy and Levenshtein distance from the gold inflection among all submissions including those using external resources.

An intermediate step in the linguistic analysis of an under-documented language is to find and organize inflected forms that are attested in natural speech. From this data, linguists generate unseen inflected word forms in order to test hypotheses about the language’s inflectional patterns and to complete inflectional paradigm tables. To get the data linguists spend many hours manually creating interlinear glossed texts (IGTs). We introduce a new task that speeds this process and automatically generates new morphological resources for natural language processing systems: IGT-to-paradigms (IGT2P). IGT2P generates entire morphological paradigms from IGT input. We show that existing morphological reinflection models can solve the task with 21% to 64% accuracy, depending on the language. We further find that (i) having a language expert spend only a few hours cleaning the noisy IGT data improves performance by as much as 21 percentage points, and (ii) POS tags, which are generally considered a necessary part of NLP morphological reinflection input, have no effect on the accuracy of the models considered here.

In supervised learning of morphological patterns, the strategy of generalizing inflectional tables into more abstract paradigms through alignment of the longest common subsequence found in an inflection table has been proposed as an efficient method to deduce the inflectional behavior of unseen word forms. In this paper, we extend this notion of morphological ‘paradigm’ from earlier work and provide a formalization that more accurately matches linguist intuitions about what an inflectional paradigm is. Additionally, we propose and evaluate a mechanism for learning full human-readable paradigm specifications from incomplete data—a scenario when we only have access to a few inflected forms for each lexeme, and want to reconstruct the missing inflections as well as generalize and group the witnessed patterns into a model of more abstract paradigmatic behavior of lexemes.