In 2026, we held the fourth iteration of the SHROOM Shared Task series: SHROOM-Visions (Shared-task on Hallucinations and Related Observable Overgeneration Mistakes in Vision language models), which is hosted at the UncertaiNLP Workshop co-located with EMNLP 2026. Following the success of the 2024 and 2025 tasks, this time we aim to tackle hallucinations through a model-agnostic detection task focused on large vision-language models. Building on the recently introduced SHEEP dataset, designed for long-term evaluation across model generations, the task invites participants to detect and classify fine-grained hallucination spans in image-conditioned text generation (VQA, image captioning, etc.). The evaluation uses a five-class taxonomy of hallucinations spanning four languages: Chinese, English, French, and Italian. The shared task generated strong interest in the NLP community worldwide, with 27 teams contributing 600+ system submissions. The best systems achieve average scores of 0.58 in character-level correlation, 0.46 in label-conditioned correlation, and 0.51 in intersection-over-union (IoU) across four languages, outperforming the baselines by 30–40 points.

Hallucinations are a persistent challenge in natural language generation, including data-to-text. van Deemter (2024) introduced a framework based on the relation of logical consequence (“follows from”), which divides all data-to-text hallucinations into seven disjoint categories. We examine whether human annotators and large language models are able to apply the framework, in two data-to-text domains. Results suggest that the framework is applicable, although there are significant domain-dependent variations, as well as discrepancies between human and model judgments. We also uncover several issues that should inform future work on hallucination.

We investigate how synonyms evolve over time, focusing on two apparently contradictory tendencies: the Law of Differentiation, whereby synonyms diverge in meaning, and the Law of Parallel Change, whereby they co-evolve. We develop a framework that breaks down the laws into a fine-grained taxonomy, which we use to construct a diachronic dataset based on the Oxford English Dictionary. We rely on definition embeddings and the use of LLM-as-a-Judge to link synonyms. We validate our methodology by performing a human evaluation on a subset of the constructed dataset. Finally, we populate the taxonomy and conduct extensive analyses across different time spans to investigate the predominance of each law in English over time. We find that while the Law of Differentiation is larger in absolute numbers, it is driven by the general tendency for meanings to disappear from the language rather than by a property unique to synonymy.

In an age of rapid model turnover, how do we make hallucination evaluation more perennial? We explore whether human-written hallucination samples could take the place of model-generated hallucinations, in order to make benchmarking detection independent of particular models. To this end, we construct SHEEP, a dataset of 1,600 human-written samples, spanning four languages (Chinese, English, French, Italian), and 18,400 samples from five vision-and-language models, all annotated for hallucinations using a fine-grained span-level labeling scheme. We find that human-written samples result in higher agreement and allow greater control of dataset contents, while remaining distributionally similar to samples derived from vision-and-language samples and providing a reasonable portrayal of detection capabilities — suggesting that human data is a viable substitute for model-based hallucination benchmarks.

When you use a large language model (LLM) in your research, you often need to formulate a prompt to elicit some relevant output from the LLM. This step is challenging since (1) LLMs are known to be brittle and their results may vary drastically between different prompts; and (2) for any given task, there are infinitely many possible prompts. Thus we end up with the following problem: If you cannot try out infinitely many prompts, how do you justify your selected prompt? This article discusses different ways in which authors may motivate their choices, and urges authors to consider how their prompting strategy might be justified.

The inaugural YNLG workshop served as a platform for community building and knowledge transfer among early-career researchers. This document synthesizes the core insights from the workshop and presents a comprehensive summary of advice from the Program Committee on navigating research methodology, publishing, and mental health. Furthermore, it highlights key takeaways from a panel of experts on mentorship, the evolving landscape of academia versus industry, and the future challenges of evaluation in NLG.

Hallucinations are one of the most pressing challenges for large language models (LLMs). While numerous methods have been proposed to detect and mitigate them automatically, human evaluation continues to serve as the gold standard. However, these human evaluations of hallucinations show substantial variation in definitions, terminology, and evaluation practices. In this paper, we survey 64 studies involving human evaluation of hallucination published between 2019 and 2024, to investigate how hallucinations are currently defined and assessed. Our analysis reveals a lack of consistency in definitions and exposes several concerning methodological shortcomings. Crucial details, such as evaluation guidelines, user interface design, inter-annotator agreement metrics, and annotator demographics, are frequently under-reported or omitted altogether.

Data-to-text natural language generation (NLG) models may produce outputs that closely mirror the structure of their input. We introduce formulaicness as a measure of the output-to-input structural resemblance, proposing it as an enhancement for reference-less naturalness evaluation. Focusing on logic-to-text generation, we construct a dataset and train a regressor to predict formulaicness scores. We collect human judgments on naturalness and examine how incorporating formulaicness into existing metrics affects alignment with these judgments.

Human evaluations lay at the heart of evaluations within the field of Natural Language Processing (NLP). Seen as the “golden standard” of evaluations, questions are being asked on whether these evaluations are both reproducible and repeatable. One overlooked aspect is the design choices made by researchers when designing user interfaces (UIs). In this paper, four UIs used in past NLP human evaluations are assessed by UX experts, based on standardized human-centered interaction principles. Building on these insights, we derive several recommendations that the NLP community should apply when designing UIs, to enable more consistent human evaluation responses.

How does the word analogy task fit in the modern NLP landscape? Given the rarity of comparable multilingual benchmarks and the lack of a consensual evaluation protocol for contextual models, this remains an open question. In this paper, we introduce MATS: a multilingual analogy dataset, covering forty analogical relations in six languages, and evaluate human as well as static and contextual embedding performances on the task. We find that not all analogical relations are equally straightforward for humans, static models remain competitive with contextual embeddings, and optimal settings vary across languages and analogical relations. Several key challenges remain, including creating benchmarks that align with human reasoning and understanding what drives differences across methodologies.

Some applications of artificial intelligence make it desirable that logical formulae be converted computationally to comprehensible natural language sentences. As there are many logical equivalents to a given formula, finding the most suitable equivalent to be used as input for such a “logic-to-text” generation system is a difficult challenge. In this paper, we focus on the role of brevity: Are the shortest formulae the most suitable? We focus on propositional logic (PL), framing formula minimization (i.e., the problem of finding the shortest equivalent of a given formula) as a Quantified Boolean Formulae (QBFs) satisfiability problem. We experiment with several generators and selection strategies to prune the resulting candidates. We conduct exhaustive automatic and human evaluations of the comprehensibility and fluency of the generated texts. The results suggest that while, in many cases, minimization has a positive impact on the quality of the sentences generated, formula minimization may ultimately not be the best strategy.

Logic-to-text generation is an important yet underrepresented area of natural language generation (NLG). In particular, most previous works on this topic lack sound evaluation. We address this limitation by building and evaluating a system that generates high-quality English text given a first-order logic (FOL) formula as input. We start by analyzing the performance of Ranta (2011)’s system. Based on this analysis, we develop an extended version of the system, which we name LoLa, that performs formula simplification based on logical equivalences and syntactic transformations. We carry out an extensive evaluation of LoLa using standard automatic metrics and human evaluation. We compare the results against a baseline and Ranta (2011)’s system. The results show that LoLa outperforms the other two systems in most aspects.

GECko+ : a Grammatical and Discourse Error Correction Tool We introduce GECko+, a web-based writing assistance tool for English that corrects errors both at the sentence and at the discourse level. It is based on two state-of-the-art models for grammar error correction and sentence ordering. GECko+ is available online as a web application that implements a pipeline combining the two models.