Human annotation is the empirical foundation of much NLP research, from dataset construction to model evaluation, but papers often leave unclear who produced the annotations and how the annotation process was controlled. We provide the first large-scale, task-level audit of human annotation reporting across major NLP venues, asking which annotation details are documented, which are missing, and how reporting varies across time, topic, venue, and intended use of human judgment. We introduce a unified taxonomy of annotation-reporting practices and validate an LLM-assisted extraction pipeline against Annotated_gold, a human-adjudicated gold standard of 41 papers and 72 annotation tasks, where the best model reaches human-comparable agreement with adjudicated labels, with Krippendorff’s alpha = 0.606 versus 0.585 for human–human agreement. Using this pipeline, we construct Annotated_llm, a dataset covering ACL-venue papers from 2018–2025, with 2,667 extracted annotation tasks from 1,603 papers, and find that papers frequently report operational details such as recruitment strategies, annotator expertise, and annotation volume, but often omit details needed to assess annotation validity, including training, language proficiency, compensation, socio-demographics, adjudication, and agreement values, especially in model-evaluation studies. Our results show that annotation reporting in NLP has improved over time but remains uneven, and they establish a scalable framework and bare-minimum reporting recommendations for making human annotation more reliable, reproducible, and interpretable.

Emotional appeals (EAs) are often treated as fallacious in computational argumentation, yet they may be justified when warranted by the context. We introduce the task of predicting EA justification and present the first human-annotated dataset for this task, covering 1,176 argument–context pairs. Grounded in philosophical literature, our framework evaluates EA justification along logical and emotional dimensions. Human annotations show that justification criteria form a shared latent judgment space, with evidential proportionality and normative acceptability most strongly associated with overall justification. We further find that EAs without logical support are not always judged fully unjustified, and stronger EAs are not necessarily judged less justified. Finally, LLM experiments show that EA justification prediction remains challenging.

We revisit the under-explored phenomenon of syntactic complexity convergence in dialogue. Extending prior topic-level analyses to full conversations, we examine five human-human datasets and a set of human-LLM dialogues. We integrate and evaluate a multidimensional set of syntactic complexity metrics to quantify structural variation in dialogue. Our analyses reveal robust syntactic complexity convergence across human-human datasets, especially in task-oriented interactions, and statistical tests confirm that these effects are not due to chance. In contrast, human-LLM dialogues show no significant convergence under individual metrics; however, linear-CKA uncovers a distinct cross-metric convergence pattern. Motivated by these differences, we use syntactic complexity as a feature for detecting LLM-involved dialogue, achieving 96% accuracy. Our classification results demonstrate its potential without relying on likelihood-based signals such as surprisal, which require a language model as the estimator.

Logical fallacies are common in public communication and can mislead audiences; fallacious arguments may still appear convincing despite lacking soundness, because convincingness is inherently subjective. We present the first computational study of how emotional framing interacts with fallacies and convincingness, using large language models (LLMs) to systematically change emotional appeals in fallacious arguments. We benchmark eight LLMs on injecting emotional appeal into fallacious arguments while preserving their logical structures, then use the best models to generate stimuli for a human study. Our results show that LLM-driven emotional framing reduces human fallacy detection in F1 by 14.5% on average. Humans perform better in fallacy detection when perceiving enjoyment than fear or sadness, and these three emotions also correlate with significantly higher convincingness compared to neutral or other emotion states. Our work has implications for AI-driven emotional manipulation in the context of fallacious argumentation.

Emotions have been shown to play a role in argument convincingness, yet this aspect is underexplored in the natural language processing (NLP) community. Unlike prior studies that use static analyses, focus on a single text domain or language, or treat emotion as just one of many factors, we introduce a dynamic framework inspired by manipulation checks commonly used in psychology and social science; leveraging LLM-based manipulation checks, this framework examines the extent to which perceived emotional intensity influences perceived convincingness. Through human evaluation of arguments across different languages, text domains, and topics, we find that in over half of cases, human judgments of convincingness remain unchanged despite variations in perceived emotional intensity; when emotions do have an impact, they more often enhance rather than weaken convincingness.We further analyze whether 11 LLMs behave like humans in the same scenario, finding that while LLMs generally mirror human patterns,they struggle to capture nuanced emotional effects in individual judgments.

Large Language Models (LLMs) have revolutionized various Natural Language Generation (NLG) tasks, including Argument Summarization (ArgSum), a key subfield of Argument Mining. This paper investigates the integration of state-of-the-art LLMs into ArgSum systems and their evaluation. In particular, we propose a novel prompt-based evaluation scheme, and validate it through a novel human benchmark dataset. Our work makes three main contributions: (i) the integration of LLMs into existing ArgSum systems, (ii) the development of two new LLM-based ArgSum systems, benchmarked against prior methods, and (iii) the introduction of an advanced LLM-based evaluation scheme. We demonstrate that the use of LLMs substantially improves both the generation and evaluation of argument summaries, achieving state-of-the-art results and advancing the field of ArgSum. We also show that among the four LLMs integrated in (i) and (ii), Qwen-3-32B, despite having the fewest parameters, performs best, even surpassing GPT-4o.

Natural Language Generation (NLG), and more generally generative AI, are among the currently most impactful research fields. Creative NLG, such as automatic poetry generation, is a fascinating niche in this area. While most previous research has focused on forms of the Turing test when evaluating automatic poetry generation — can humans distinguish between automatic and human generated poetry — we evaluate the diversity of automatically generated poetry (with a focus on quatrains), by comparing distributions of generated poetry to distributions of human poetry along structural, lexical, semantic and stylistic dimensions, assessing different model types (word vs. character-level, general purpose LLMs vs. poetry-specific models), including the very recent LLaMA3-8B, and types of fine-tuning (conditioned vs. unconditioned). We find that current automatic poetry systems are considerably underdiverse along multiple dimensions — they often do not rhyme sufficiently, are semantically too uniform and even do not match the length distribution of human poetry. Our experiments reveal, however, that style-conditioning and character-level modeling clearly increases diversity across virtually all dimensions we explore. Our identified limitations may serve as the basis for more genuinely diverse future poetry generation models.

Recently proposed BERT-based evaluation metrics for text generation perform well on standard benchmarks but are vulnerable to adversarial attacks, e.g., relating to information correctness. We argue that this stems (in part) from the fact that they are models of semantic similarity. In contrast, we develop evaluation metrics based on Natural Language Inference (NLI), which we deem a more appropriate modeling. We design a preference-based adversarial attack framework and show that our NLI based metrics are much more robust to the attacks than the recent BERT-based metrics. On standard benchmarks, our NLI based metrics outperform existing summarization metrics, but perform below SOTA MT metrics. However, when combining existing metrics with our NLI metrics, we obtain both higher adversarial robustness (15%–30%) and higher quality metrics as measured on standard benchmarks (+5% to 30%).

We consider the end-to-end abstract-to-title generation problem, exploring seven recent transformer based models (including ChatGPT) fine-tuned on more than 30k abstract-title pairs from NLP and machine learning (ML) venues. As an extension, we also consider the harder problem of generating humorous paper titles. For the latter, we compile the first large-scale humor annotated dataset for scientific papers in the NLP/ML domains, comprising 2.6k titles. We evaluate all models using human and automatic metrics. Our human evaluation suggests that our best end-to-end system per-forms similarly to human authors (but arguably slightly worse). Generating funny titles is more difficult, however, and our automatic systems clearly underperform relative to humans and often learn dataset artefacts of humor. Finally, ChatGPT, without any fine-tuning, performs on the level of our best fine-tuned system.

Reproducibility is of utmost concern in machine learning and natural language processing (NLP). In the field of natural language generation (especially machine translation), the seminal paper of Post (2018) has pointed out problems of reproducibility of the dominant metric, BLEU, at the time of publication. Nowadays, BERT-based evaluation metrics considerably outperform BLEU. In this paper, we ask whether results and claims from four recent BERT-based metrics can be reproduced. We find that reproduction of claims and results often fails because of (i) heavy undocumented preprocessing involved in the metrics, (ii) missing code and (iii) reporting weaker results for the baseline metrics. (iv) In one case, the problem stems from correlating not to human scores but to a wrong column in the csv file, inflating scores by 5 points. Motivated by the impact of preprocessing, we then conduct a second study where we examine its effects more closely (for one of the metrics). We find that preprocessing can have large effects, especially for highly inflectional languages. In this case, the effect of preprocessing may be larger than the effect of the aggregation mechanism (e.g., greedy alignment vs. Word Mover Distance).

This paper describes our contribution to the Shared Task ReproGen by Belz et al. (2021), which investigates the reproducibility of human evaluations in the context of Natural Language Generation. We selected the paper “Generation of Company descriptions using concept-to-text and text-to-text deep models: data set collection and systems evaluation” (Qader et al., 2018) and aimed to replicate, as closely to the original as possible, the human evaluation and the subsequent comparison between the human judgements and the automatic evaluation metrics. Here, we first outline the text generation task of the paper of Qader et al. (2018). Then, we document how we approached our replication of the paper’s human evaluation. We also discuss the difficulties we encountered and which information was missing. Our replication has medium to strong correlation (0.66 Spearman overall) with the original results of Qader et al. (2018), but due to the missing information about how Qader et al. (2018) compared the human judgements with the metric scores, we have refrained from reproducing this comparison.