Topics

We welcome proposals for papers on all work addressing the issue of LLM evaluation.

This includes research on:

•     evaluation of complete systems, including RAG, or its components (e.g., retrievers, chunkers, etc.)
•     evaluation of foundation models, fine-tuned models, or complete systems (e.g., RAG)
•     the creation or adaptation of benchmarks for French or other languages of interest, whether well-endowed or poorly endowed, in general or specialized domains, or for noisy or non-standard languages (e.g., social networks, voice commands, etc.)
•     evaluation on NLP tasks (translation, summarization, information extraction, etc.)
•     the adaptation of existing evaluation methodologies to generative systems
•     ethical dimensions, bias, privacy, cultural or legislative alignment
• performance dimensions in terms of computation time, memory, energy efficiency
• evaluation with users, ergonomics, cognitive aspects
• evaluation of multimodal models (e.g., text-image, text-speech, etc.).
•  ...

The proceedings of the workshop will be published.


Format and type of articles


Several types of articles are accepted:

    new contributions,
    state of the art,
    work in progress,
    short/translated versions of articles accepted at a major conference.


The length of articles is flexible, between 4 and 10 pages, excluding references.

As reviews are conducted on a double-blind basis, submissions must be anonymized. The style sheet is that of the TALN 2026 conference. Authors are invited to comply with best practices in the field: https://aclrollingreview.org/responsibleNLPresearch/ Articles should be submitted on this page (TBC).

Call schedule

Submission: TBC Feedback to authors: TBC Final version: TBC