International Workshop on Structures to Stories: Multilingual NLP for Semantic Structured Data (IWSS) – co-located with The 25th International Semantic Web Conference

Despite recent advances in LLM-based pipelines, the core multilingual Natural Language Processing (NLP) and Natural Language Generation (NLG) challenges of mapping between structured data and natural language, especially in low-resource and long-tail settings, remain largely unsolved and under-represented within the Semantic Web community.

The International Workshop on Structures to Stories: Multilingual NLP for Semantic Structured Data (IWSS) brings together researchers and practitioners advancing multilingual methods to (i) extract structured representations from text and (ii) generate faithful answers, verbalisations, explanations, and narratives grounded in structured evidence.

IWSS emphasises robust extraction, data-grounded generation, and realistic multilingual evaluation, with particular interest in under-represented languages, long-tail entities, and culturally specific concepts.

We invite full and short papers, position and negative-result papers, datasets, system demos, and resource papers. Topics include faithful multilingual verbalisation and summarisation, entity extraction and linking, schema/ontology alignment, and low-resource benchmarks and evaluation, among others.

Overview

Date: 25-26 October, 2026
Location: Bari, Italy (co-located with The 25th International Semantic Web Conference)
Format: Half-day
Contact: iwss@outlook.es

News

Update Call for Papers published.
Update Important dates announced.
Update Program coming soon.

Submission

Easychair submission
CEUR-WS Proceedings Template

Topics

Topics of interest include (but not limited to):

Multilingual data-to-text (grounded verbalisation/explanations) and text-to-data (Information Extraction/Semantic Parsing, Entity Linking/normalisation)
Cross-lingual schema/ontology alignment and lexicalisation
Faithfulness/robustness for structured data NLP/NLG
Low-resource benchmarks, datasets, and evaluation for structured data
Limitations of current metrics in evaluating NLG outputs
Analysis of biases introduced in data-to-text and debiasing solutions
Data-to-text solutions for data sources containing long-tail entities

Important Dates

All deadlines are Anywhere on Earth (AoE) unless stated otherwise.

Milestone	Date
Paper submission deadline	July 24, 2026
Notification of acceptance	August 21, 2026
Camera-ready deadline	August 30, 2026
Workshop date	October 25-26, 2026

Call for Papers

Submission Instructions

Submission site: EasyChair
Format: PDF, using the CEUR-WS (CEURART) template
Page limits (references excluded):
- Full research papers: 8–12 pages
- In-Use and Experience papers: 8–12 pages
- Short research papers: 4–6 pages
- System/Demo/Position papers: 4–6 pages
Repositories: Authors are welcome and encouraged to submit accompanying repositories (e.g., code, data, models) in line with FAIR principles. /.
Template: CEUR-WS Overleaf template (CEURART)

Review & Publication

Single-blind review process.
Camera-ready instructions will be shared with accepted authors.
Proceedings shall be submitted to CEUR-WS.org for online publication under the Creative Commons Attribution 4.0 (CC BY 4.0) license. The final camera-ready version must include the mandatory CC BY 4.0 license notice on the first page, as required by CEUR-WS.
Papers must follow CEURART instructions, including all mandatory elements. This includes the required Declaration on Generative AI (where applicable).
At least one of the authors of each accepted paper must attend the workshop in person to present the work.
The Author Agreement must be physically signed in ink by the contact author and provided as a scanned copy (PDF or JPG) to the organizers by the camera-ready deadline:
- Author Agreement (no copyrighted third-party material included)
- Author Agreement (copyrighted third-party material included)