SIGHUM 2027 Workshop (former LaTeCH-CLfL)

Dear all, we're happy to announce our CfP for the EACL Workshop SIGHUM 2027 (a.k.a. LaTeCH-CLfL)
The 11th Joint SIGHUM Workshop on Computational Linguistics
for Cultural Heritage, Social Sciences, Humanities and Literature
to be held at EACL in March 2027 in Athens, Greece as a one-day workshop
Organizers: Diego Alves, Yuri Bizzoni, Stefania Degaetano-Ortlieb,
Anna Kazantseva, Janis Pagel, Stan Szpakowicz
First Call for Papers (with apologies for cross-posting)
SIGHUM 2027 is the eleventh in a series of meetings for NLP researchers who work
with data from the broadly understood arts, humanities and social sciences, and for
specialists in those disciplines who apply NLP techniques in their work. The workshop
continues a long tradition of annual events which also host the SIGHUM business
meetings.
Workshop site
https://sighum.wordpress.com/sighum-2027/
Important dates
Submission deadline: December 15th, 2026
Notification of acceptance: January 20th, 2027
Camera-ready paper due: January 29th, 2027
Description
The community of the broadly understood Digital Humanities (DH) has witnessed remarkable
growth and transformation, fuelled by the rapid advancements in NLP. There
is a steady interest in, and a high demand for, NLP methods of semantic and structural
annotation, intelligent linking, discovery, querying, cleaning and visualization of primary
and secondary data. Even so, the heterogeneous landscape of the DH – with their diverse,
often multilingual or multimodal sources – can be a challenge for NLP. Consider,
for example, the growing interest in historical language data and in under-resourced
languages.
There are unique obstacles in developing comprehensive language models in aid of the
linguistic diversity in DH. The handling of noisy and non-standard data, and the need
for domain adaptation and intensive annotation, stay at the forefront of research effort
in the community. The literary studies, which have witnessed substantial progress in
the application of NLP methods, bring their own similar problems. Navigating forms
of creative expression requires more than the typical information-seeking tools. A case
in point might be the study of literature of a certain period, author or sub-genre, the
recognition of certain literary devices, or the quantitative analysis of poetry.
The Large Language Models (LLMs) expand the DH toolkit. There is support for automatic
text cleaning and annotation, creation of semantic resources, analysis of narrative,
genre and literary style, and linking information across sources. LLMs can support historical
or low-resource languages, particularly when complemented with domain-specific
fine-tuning and careful evaluation. One must note, however, that even with careful adaptation,
curation and attention to interpretability, LLM outputs remain prone to errors,
bias and lack of transparency; rigorous assessment is required to ensure their suitability
for scholarly research.
There is growing emphasis on the importance of explanation in NLP models. That
applied equally to DH, whose various domains enjoy the effect of NLP. Transparency and
clarity of the results are critical if one is to accept the processed data and gain valuable
insights. That is why one must carefully consider a balance between raw performance
scores and interpretability, in keeping with the specific research objectives.
For many years now, this broad research context has drawn together NLP experts,
data specialists and researchers in Digital Humanities who work in and across their domains.
Our long-standing series of workshops has shown that cross-disciplinary exchange
supports work in the Humanities, Social Sciences and Cultural Heritage communities. It
encourages the Computational Linguistics community to build rich and effective tools,
and – above all – explainable models.
Topics
Our workshops attract original work on a wide variety of topics, including – but as usual
not restricted to – these:
• adaptation of NLP tools to Cultural Heritage, Social Sciences, Humanities and
literature;
• automatic error detection and cleaning of textual data;
• complex annotation plans, tools and interfaces;
• creation (fully- or semi-automatic) of semantic resources;
• creation and analysis of social networks of literary characters;
• discourse and narrative analysis/modelling, notably in literature;
• emotion analysis for the humanities and for literature;
• generation of literary narrative, dialogue or poetry;
• identification and analysis of literary genres;
• information/knowledge modelling in the Humanities, Social Sciences and Cultural
Heritage;
• interpretability of large language models output for DH-related tasks (explainable
AI);
• linkage and retrieval of information from different sources, media, and domains;
• low-resource and historical language processing;
• modelling dialogue literary style for generation;
• profiling and authorship attribution;
• search for scientific or scholarly literature;
• work with linguistic variation and non-standard or historical use of language.
Information for authors
We invite papers on original, unpublished work in the topic areas of the workshop.
We will consider long papers, short papers and system descriptions (demos). We also
welcome position papers.
A long paper, which presents completed work, may have up to eight (8) pages of
content plus additional pages of references (just two, please, if possible). The final
camera-ready versions of accepted long papers will get one additional page of content
(up to 9 pages), so that reviewers’ comments can be taken into account.
A short paper or a demo, which presents work in progress or the description of a
system, may have up to four (4) pages of content plus additional pages of references
(one if you can, please). Upon acceptance, short papers will get five (5) content pages
in the proceedings.
A position paper – clearly marked as such – should not exceed eight (8) pages including
references.
All submissions are to follow the *ACL paper styles (for LaTeX / Overleaf and MS
Word) available at https://github.com/acl-org/acl-style-files. Papers will be
submitted electronically, only in PDF, via the SIGHUM 2027 submission website on the
SoftConf pages – we will post the link as soon as we have it.
Reviewing will be double-blind. Please do not include the authors’ names and affiliations,
or any references to Web sites, project names, acknowledgments, and so on –
anything that immediately reveals the authors’ identity. Please keep references to your
own work at a reasonable minimum, and do not include anonymous citations.
In general accordance with the Multiple Submission Policy at EACL 2027, we will not
consider any paper simultaneously under review in a journal or at another conference.

During the review period, papers submitted to our workshop cannot also be submitted
elsewhere.
The reviewers will look out for any signals of the use of AI in the writing of your
paper. In particular, any hallucinated bibliographic reference means immediate
rejection.
Non-archival poster presentation of ACL Findings papers
For non-archival poster presentation at the workshop, we will consider papers accepted as
Findings of ACL. To qualify, a paper must align very closely with the scope of SIGHUM.
We aim to bring together research relevant to language technologies for the Humanities
and Cultural Heritage, otherwise distributed across ACL areas. That should foster
exchange in our research community. We will base selection on the already reviewed
papers’ thematic relevance to the workshop and on the availability of (few) presentation
slots. Our decisions will be final.