[Event at CIG] [Deadline extension][CFP] 2nd International Workshop on Generative AI for Textual Document Analysis (GENAIDOC)

[rafika boutalbi]

*2nd International Workshop on Generative AI for Textual Document Analysis
(GENAIDOC)*

*https://sites.google.com/view/genaidoc-workshop-fllm-2025*
<https://genaidoc-workshop-fllm-2025/>

*As part of the 3rd International Conference on Foundation and Large
Language Models (FLLM2025)*

*https://fllm-conference.org/2025/index.php*
<https://fllm-conference.org/2025/index.php>

*November 25 to November 28, 2025, Vienna, Austria*

*Context:*

Nowadays, the volume of textual data being generated is unprecedented. From
social media posts, news articles, and academic papers to customer reviews,
emails, and business documents, the sheer quantity of text data is growing
exponentially. Traditional methods of analyzing this vast amount of data
often fall short in terms of scalability, accuracy, and efficiency. In this
context, Generative AI (GenAI) is revolutionizing the field of Natural
Language Processing (NLP) by enabling the creation of highly sophisticated
Large Language Models (LLMs) that can generate, understand, and manipulate
human language. GenAI models like GPT-4 and BERT are at the forefront of
these advancements from chatbots to automated content creation. This
workshop aims to provide participants with a deep understanding of LLMs,
its applications in NLP, and the ethical considerations involved. GenAI
models are designed to handle and process enormous datasets, making them
ideal for textual document analysis. These models leverage advanced machine
learning techniques to understand, interpret, and generate human-like text,
allowing for more nuanced and comprehensive analysis. By using LLMs, we can
uncover insights and patterns that would be impossible to detect using
conventional methods.

*Objective:*

This workshop is designed to provide a comprehensive understanding of how
LLMs can be leveraged for textual document analysis. Participants will gain
hands-on experience and theoretical knowledge about the applications,
capabilities, and limitations of GenAI models in the context of analyzing
textual data. The workshop will cover various techniques and tools,
practical implementation, and the latest advancements in the field. The
GENAIDOC workshop aims to bring together an area for experts from industry,
science, and academia to exchange ideas and discuss ongoing research in
natural language processing and GenAI for textual document analysis.

Novelty for this edition:

After the success of GENAIDOC 2024
<https://sites.google.com/view/genaidoc-workshop-fllm-2024/home> at
FLLM’24,  in this edition of GENAIDOC will further explore a fast-emerging
yet underformalized area at the intersection of large language models
(LLMs), document understanding, and multimodal AI. The proposed extension
focuses on the design of effective prompting strategies for extracting,
interpreting, and aligning textual and visual information from real-world
documents such as scanned PDFs, structured forms, and tabular data.

The new focus will delve into the design of prompting strategies that
enable effective extraction, interpretation, and alignment of textual and
visual elements from scanned documents, PDFs, structured forms, and tabular
data. This includes analyzing how prompts influence the performance of LLMs
when processing OCR-based content, determining the best approaches for
handling visual structures such as tables and layout elements, and studying
how textual and visual modalities can be integrated or separated during
inference. The topic will also cover the development of standardized prompt
templates adapted to industrial contexts such as invoice and document
processing. This direction aims to combine technical depth with real-world
applicability, enhancing both academic and practical contributions to the
field.

Topics of interests:


This workshop invites submissions with high-quality works that are related,
but are not limited, to the topics below:

   -
      -

      Prompts to extract textual information
      -

      Prompts to extract visual information
      -

      Prompts to extract data from tables
      -

      Prompts for specific documents type
      -

      Prompts to classify documents
      -

      Text classification
      -

      Automatic document summarization
      -

      Automatic machine translation
      -

      Sentiment analysis
      -

      Text generation
      -

      Deep learning for NLP
      -

      Reinforcement Learning for NLP
      -

      Unsupervised Learning for NLP
      -

      Speaker identification
      -

      Speech recognition
      -

      Speech to Text
      -

      Text detection and recognition from images
      -

      Question Answering systems
      -

      Transfer Learning for NLP
      -

      Active Learning for NLP
      -

      Real-life and industrially relevant NLP applications
      -

         Email filtering
         -

         invoice information extraction
         -

         News generation
         -

         Meeting analysis
         -

         CVs analysis and classification


Submission:

Papers submitted for review should conform to IEEE specifications.
Manuscript templates can be downloaded from IEEE website
<https://streaklinks.com/CBGLsQgm5i0NAZelDgCJIkDi/https%3A%2F%2Fwww.ieee.org%2Fconferences%2Fpublishing%2Ftemplates.html>.
The maximum length of papers is 8 pages. All the papers will go through the
double-blind peer review process. Authors’ names and affiliations should
not appear in the submitted paper. Authors’ prior work should be cited in
the third person. Authors should also avoid revealing their identities
and/or institutions in the text, figures, links, etc.

Authors should also ensure that their identity is not revealed indirectly
by citing their previous work in the third person and omitting
acknowledgments until the camera-ready version. Papers have to be submitted via
the workshop's EasyChair
<https://easychair.org/conferences/?conf=fllm2025> submission
page.

Please include in the paper title "Full paper: Title" or "Short paper:
Title" to precise the contribution type. At least one author of each
accepted paper must register for the workshop, in order to present the paper
. For further instructions, please refer to the FLLM 2025 page
<https://fllm-conference.org/2025/index.php>.

*Important dates: *

   -

   Submission Deadline: August 31, 2025  *September 21st**, 2025*
   -

   Decisions Announced: September 30, 2025 *October 10th**, 2025*
   -

   Camera Ready Deadline: October 08, 2025 *October 25th, 2025*

Workshop: To be announced

*Publication*:

Accepted papers will be submitted to IEEEXplore for possible publication.

Workshop Chairs

Rim Hantach <http://rim.hantach%40gmail.com%20%3Crim.hantach@gmail.com%3E;/>,
Engie, France

Rafika Boutalbi
<https://streaklinks.com/CBGLsQggIIcXJ_PtQgsr0IDJ/http%3A%2F%2Frafika.boutalbi%40univ-amu.fr%2F>,
Aix-Marseille University, France

*Karima Boutalbi* <karima.boutalbi at cgedim.com>, Cegedim Business Services,
France