arXiv subjects

Nicolas Turenne

Publications and source records attributed to Nicolas Turenne.

10 recordsLinked to original sources

arXiv2026

Automated pipeline for herbarium label digitization

Digitized herbarium collections, now comprising over 100 million freely accessible specimen images, have become a critical resource for addressing fundamental questions in ecology and evolutionary biology. Yet the rich metadata encoded in herbarium labels (collector identities, geographic localities, collection dates, and ecological observations) remains largely inaccessible at scale, constraining both biodiversity informatics and the construction of specimen-specific image-text corpora for multimodal AI. We present HERBIOME, a modular end-to-end pipeline for automated herbarium label digitization, integrating YOLOv8-based component detection, CRAFT Hezar word-level text localization, fine-tuned TrOCR for recognition of mixed handwritten and printed text, and GPT-4o Mini for semantic metadata structuring into standardized fields. TrOCR was trained on a multi-source dataset combining general transcription corpora (CREMMA-AN, PictoCatalogs) with herbarium-specific data (RéColNat), achieving a Character Error Rate of 4.05-4.10%. End-to-end evaluation on 450 French herbarium specimens, using a dual-metric framework of Maximum Window Similarity (MWS: 0.614-0.618) and Semantic Metadata Accuracy (SMA: 0.440-0.445), reveals that hybrid training strategies improve semantic fidelity while random sampling maximizes surface similarity, with taxonomic fields remaining the principal bottleneck. By automating the extraction of structured metadata from complex, heterogeneous labels, HERBIOME reduces transcription burden, enables the construction of paired image-text datasets that faithfully capture specimen individuality, which is a prerequisite for next-generation multimodal biodiversity AI systems.

cs.CV↗

arXiv2026

A Generative Grammar Underlying the Voynich Manuscript, the Pastiche Hypothesis: Evidence from Large Language Models

Background: The Voynich Manuscript is a fifteenth-century codex written in an unknown script whose content remains undeciphered. Previous studies suggest that its statistical properties resemble those of natural languages, while its illustrations - primarily plants - recall medieval herbals. Methods: We present a multidisciplinary analysis combining probabilistic modeling, phonetic decomposition, rare-event detection, and multimodal image analysis, based on a newly transliterated corpus. Word- and letter-level distributions are modeled using position-dependent probabilistic grammars, while phonetic patterns are compared across Indo-European, Semitic, and Asian languages. Image-text alignment methods based on large language models are applied to identify potential botanical correspondences. Results: The results indicate that Voynich symbols behave as letters rather than syllabic units, while word-length distributions resemble syllabic structures. Phonetic analyses show closer alignment with consonant-heavy languages such as Hebrew or Arabic than with Indo-European languages. Probabilistic modeling reproduces Zipf-like distributions and reveals extremely low probabilities for repeated initial-letter sequences, indicating a structured imitation of natural language. Image analysis suggests strong correspondences between Voynich plant illustrations and those found in Pseudo-Apuleius herbals from the Mediterranean tradition, consistent with an imitation of medieval medicinal books. Perspectives: These findings support the hypothesis that the Voynich Manuscript follows a structured generative system combining linguistic regularities and herbal knowledge, and demonstrate the value of integrating probabilistic and AI-assisted approaches in the analysis of historical manuscripts.

cs.CL↗

arXiv2026

An Agentic Framework Using Rules and LLMs for Embedding and Annotating Descriptive Document Layouts: A Plant Science Use Case

Background: Recent advances in information retrieval (IR) leverage both dense and sparse representations, large language models (LLMs), and specialized retrieval models to improve ranking accuracy, relevance, and cross-lingual performance. Complementary techniques such as passage indexing, document layout analysis, and semantic knowledge representation further enhance retrieval effectiveness by capturing fine-grained contextual and structural information. Emerging agentic LLM frameworks extend these capabilities by enabling planning, iterative reasoning, tool use, and multi-agent collaboration, thereby broadening applications across diverse domains. These frameworks also emphasize rigorous evaluation, ethical considerations, and trustworthiness, ensuring responsible deployment in real-world settings. We propose a modular, agent-based pipeline for botanical trait extraction. Optical character recognition (OCR) converts PDFs into machine-readable text, while segmentation and indexing organize content by genus and species. Rule-based parsers extract structured botanical traits, and ensembles of large language models (LLMs) expand trait vocabularies and resolve ambiguities. This approach ensures accurate species recognition, scalable annotation, and explainable integration of textual botanical descriptions, enabling robust and interpretable data extraction across large botanical corpora. Results: Using three regional botanical datasets, our system extracted 55,737 trait annotations across 4,961 species, averaging 9.1 traits per species. Integration of LLM-based enrichment improved coverage for 75% of traits, increasing total annotations by 59%. While the choice of OCR engine had a minor effect on species recognition, overall annotation counts remained stable, demonstrating the robustness, scalability, and reliability of the pipeline for large-scale botanical trait extraction.

cs.AI↗

arXiv2015

Duplicate Detection with Efficient Language Models for Automatic Bibliographic Heterogeneous Data Integration

We present a new method to detect duplicates used to merge different bibliographic record corpora with the help of lexical and social information. As we show, a trivial key is not available to delete useless documents. Merging heteregeneous document databases to get a maximum of information can be of interest. In our case we try to build a document corpus about the TOR molecule so as to extract relationships with other gene components from PubMed and WebOfScience document databases. Our approach makes key fingerprints based on n-grams. We made two documents gold standards using this corpus to make an evaluation. Comparison with other well-known methods in deduplication gives best scores of recall (95\%) and precision (100\%).

cs.DB↗

arXiv2015

A Table-Binning Approach for Visualizing the Past

Large amounts of data are available due to low-cost and high-capacity data storage equipments. We propose a data exploration/visualization method for tabular multi-dimensional, time-varying datasets to present selected items in their global context. The approach is simple and uses a rank-based visualization and a pattern matching functionality based on temporal profiles. Ranking categories can be specified in a flexible way and are used instead of actual values (value reduction into bins) and plotting it over time in an unevenly quantized representation. Patterns that emerge are matched against a set of eight predefined temporal profiles. The graphical summarization of large-scale temporal data is proposed and applicability is tested qualitatively on about eight data sets and the approach is compared to classic line plots and SAX representation

cs.IR↗

arXiv2015

On a Possible Similarity between Gene and Semantic Networks

In several domains such as linguistics, molecular biology or social sciences, holistic effects are hardly well-defined by modeling with single units, but more and more studies tend to understand macro structures with the help of meaningful and useful associations in fields such as social networks, systems biology or semantic web. A stochastic multi-agent system offers both accurate theoretical framework and operational computing implementations to model large-scale associations, their dynamics and patterns extraction. We show that clustering around a target object in a set of associations of object prove some similarity in specific data and two case studies about gene-gene and term-term relationships leading to an idea of a common organizing principle of cognition with random and deterministic effects.

cs.CL↗

arXiv2015

Open Data Platform for Knowledge Access in Plant Health Domain : VESPA Mining

Important data are locked in ancient literature. It would be uneconomic to produce these data again and today or to extract them without the help of text mining technologies. Vespa is a text mining project whose aim is to extract data on pest and crops interactions, to model and predict attacks on crops, and to reduce the use of pesticides. A few attempts proposed an agricultural information access. Another originality of our work is to parse documents with a dependency of the document architecture.

cs.IR↗

arXiv2015

x.ent: R Package for Entities and Relations Extraction based on Unsupervised Learning and Document Structure

Relation extraction with accurate precision is still a challenge when processing full text databases. We propose an approach based on cooccurrence analysis in each document for which we used document organization to improve accuracy of relation extraction. This approach is implemented in a R package called \emph{x.ent}. Another facet of extraction relies on use of extracted relation into a querying system for expert end-users. Two datasets had been used. One of them gets interest from specialists of epidemiology in plant health. For this dataset usage is dedicated to plant-disease exploration through agricultural information news. An open-data platform exploits exports from \emph{x.ent} and is publicly available.

cs.CL↗

arXiv2015

svcR: An R Package for Support Vector Clustering improved with Geometric Hashing applied to Lexical Pattern Discovery

We present a new R package which takes a numerical matrix format as data input, and computes clusters using a support vector clustering method (SVC). We have implemented an original 2D-grid labeling approach to speed up cluster extraction. In this sense, SVC can be seen as an efficient cluster extraction if clusters are separable in a 2-D map. Secondly we showed that this SVC approach using a Jaccard-Radial base kernel can help to classify well enough a set of terms into ontological classes and help to define regular expression rules for information extraction in documents; our case study concerns a set of terms and documents about developmental and molecular biology.

cs.LG↗

arXiv2014

Clustering and Relational Ambiguity: from Text Data to Natural Data

Text data is often seen as "take-away" materials with little noise and easy to process information. Main questions are how to get data and transform them into a good document format. But data can be sensitive to noise oftenly called ambiguities. Ambiguities are aware from a long time, mainly because polysemy is obvious in language and context is required to remove uncertainty. I claim in this paper that syntactic context is not suffisant to improve interpretation. In this paper I try to explain that firstly noise can come from natural data themselves, even involving high technology, secondly texts, seen as verified but meaningless, can spoil content of a corpus; it may lead to contradictions and background noise.

cs.CL↗