Document-level text simplification involves transformations that go beyond sentence-internal edits, addressing discourse coherence, anaphora resolution, and cross-paragraph consistency. Despite advances in sentence-level simplification for high-resource languages, document-level simplification in morphologically rich, low-resource languages such as Estonian remains largely unexplored. This study presents a comprehensive evaluation of five state-of-the-art multilingual large language models (LLMs) for document-level simplification in Estonian. Three prompting strategies are examined: single-pass generation, pipeline-based modular agents, and guideline-augmented pipelines. The evaluation framework integrates automatic metrics assessing readability, semantic preservation, and discourse coherence, alongside a structured manual annotation protocol. The findings indicate that Gemini-2.0 and LLaMA-3.3 produce outputs with near-native fluency and strong meaning preservation, whereas other models display notable grammatical and semantic limitations. This work contributes novel document-level coherence metrics, evidence-based prompting strategies, and publicly available resources for reproducibility.

Large language models enable zero-shot knowledge graph extraction from text, yet evaluation at the level of complete typed graphs remains an open challenge. We present a schema-constrained evaluation framework that combines an explicit ontology of six entity types and 96 relation types with structured generation guided by schema-injected prompts. Supporting both single-step and two-step extraction modes, controlled inference settings, and repeated-run stability analysis, the framework enables systematic benchmarking of LLM-based graph construction under closed ontology constraints. Four large language models Gemini 3 Pro, GPT-5.1, Claude Opus 4.5, and Mistral 7B are evaluated on DocRED using entity and triple F1, schema adherence, and run consistency. Manual review reveals that automatic triple F1 systematically underestimates extraction quality, as a substantial portion of model-predicted triples are textually valid but absent from the incomplete gold annotations. The framework, prompts, and experimental outputs are publicly available for download and experimentation.

Current LLM ensemble frameworks focus on multi-step setups with additional modules for answer ranking, often opting for token and span analysis rather than structured outputs, leading to heavyweight architectures with potential fail states along the pipeline. Faster, lighter solutions are more vulnerable to hallucination propagation and can lack output control in more complex pipelines. This paper proposes a customisable, lightweight ensemble workflow of coordinated Large Language Models that leverages JSON-structured outputs and anonymous peer-review ranking to mitigate hallucinatory outputs and single-model failure points. The pipeline is demonstrated on a relation extraction task applied to English popular science articles, targeting four ontologically-grounded relation types (strong causation, weak causation, contrastive, and compositional), with semantic node canonicalisation and interactive, colour-coded HTML causal graphs as the final output. Performance is evaluated through an anonymous user study, achieving an average perceived accuracy of 0.778 against a human-annotated gold standard. The modular architecture supports flexible deployment across both API-based and in-house LLM setups, and the full framework is released under an open license to foster reproducibility and collaborative research.

This paper presents a method for text simplification based on two neural architectures: a neural machine translation (NMT) model and a fine-tuned large language model (LLaMA). Given the scarcity of existing resources for Estonian, a new dataset was created by combining manually translated corpora with GPT-4.0-generated simplifications. OpenNMT was selected as a representative NMT-based system, while LLaMA was fine-tuned on the constructed dataset. Evaluation shows LLaMA outperforms OpenNMT in grammaticality, readability, and meaning preservation. These results underscore the effectiveness of large language models for text simplification in low-resource language settings. The complete dataset, fine-tuning scripts, and evaluation pipeline are provided in a publicly accessible supplementary package to support reproducibility and adaptation to other languages.

This study explores the overlap between text summarization and simplification outputs. While summarization evaluation methods are streamlined, simplification lacks cohesion, prompting the question: how closely can abstractive summarization resemble gold-standard simplification? We address this by applying two BART-based BRIO summarization methods to the Newsela corpus, comparing outputs with manually annotated simplifications and achieving a top ROUGE-L score of 0.654. This provides insight into where summarization and simplification outputs converge and differ.

Researchers in Computational Linguistics build models of similarity and test them against human judgments. Although there are many empirical studies of the computational models of similarity for the English language, the similarity for other languages is less explored. In this study we are chiefly interested in two aspects. In the first place we want to know how much of the human similarity is grounded in the visual perception. To answer this question two neural computer vision models are used and their correlation with the human derived similarity scores is computed. In the second place we investigate if language influences the similarity computation. To this purpose diverse computational models trained on Estonian resources are evaluated against human judgments

A hybrid pipeline comprising rules and machine learning is used to filter a noisy web English-German parallel corpus for the Parallel Corpus Filtering task. The core of the pipeline is a module based on the logistic regression algorithm that returns the probability that a translation unit is accepted. The training set for the logistic regression is created by automatic annotation. The quality of the automatic annotation is estimated by manually labeling the training set.

The last years witnessed an increasing interest in the automatic methods for spotting false translation units in translation memories. This problem presents a great interest to industry as there are many translation memories that contain errors. A closely related line of research deals with identifying sentences that do not align in the parallel corpora mined from the web. The task of spotting false translations is modeled as a binary classification problem. It is known that in certain conditions the ensembles of classifiers improve over the performance of the individual members. In this paper we benchmark the most popular ensemble of classifiers: Majority Voting, Bagging, Stacking and Ada Boost at the task of spotting false translation units for translation memories and parallel web corpora. We want to know if for this specific problem any ensemble technique improves the performance of the individual classifiers and if there is a difference between the data in translation memories and parallel web corpora with respect to this task.

The paper describes the methodology and the tools we developed for the purpose of building a Romanian wordnet. The work is carried out within the BalkaNet European project and is concerned with wordnets for Bulgarian, Czech, Greek, Romanian, Serbian and Turkish all of them aligned via an interlingual index (ILI) to Princeton Wordnet. The wordnets structuring follows the principles adopted in EuroWordNet. In order to ensure maximal cross-lingual lexical coverage, the consortium decided to implement the same concepts, represented by a common set of ILI concepts. We describe the selection of concepts to be implemented in all the monolingual wordnets The methodologies adopted by each partner were different and they depended on the language resources and personnel available. For the Romanian wordnet,we decided that it should be based on the reference lexicographic descriptions of Romanian which we had in electronic forms: EXPD, a heavily XML annotated explanatory dictionary (developed in the previous CONCEDE project and based on the standard Explanatory Dictionary of Romanian), SYND, a published dictionary of synonyms which we keyboarded, encoded and completed with more than 4000 new synonymy sets extracted from EXPD, EnRoD, a Romanian-English dictionary, most part of it being extracted automatically from parallel corpora and further hand validated and extended. Besides these monolingual resources, as all the other members of the consortium, we had at our disposal the interlingual mapping of the Princeton Wordnet. All the above mentioned resources have been incorporated into a user-friendly system, WnBuilder, which allows for cooperative work of a large number of lexicographers. When the distributed work is put together, the synsets are validated. Several errors show up, the most frequent and difficult to solve being the case of a literal with the same sense number appearing in different synsets. We discuss reasons for such conflicts as well as their correction, supported by another utility program called WnCorrector. The full paper presents WnBuilder and WnCorrector, as well as the status of the Romanian wordnet development.