The 20th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2027) is launching a special call for research that places human language science at the heart of natural language processing (NLP) development. Taking place in Athens, Greece, from March 9 to 14, 2027, the conference has chosen "The Human in Language" as its organizing theme, signaling a shift in how the field thinks about building and evaluating AI systems that work with text.

Why Are NLP Researchers Suddenly Focused on the Human Element?

The special theme invites researchers to explore how insights from human language sciences can inform every stage of NLP work, from the moment a model is designed through to how it performs in real-world settings. This includes model development, how data is collected and labeled, how systems are evaluated, and ultimately how they interact with users and are deployed in practice.

The conference organizers are also welcoming work that flips the equation: using NLP methods and modern language models to advance our understanding of language, cognition, communication, and the brain itself. The central goal remains strengthening the role of traditional language sciences as a source of scientific grounding, methodological guidance, and conceptual insight for NLP itself.

What Topics Will the Conference Cover?

EACL 2027 maintains a broad technical program spanning dozens of research areas. The conference welcomes submissions across a comprehensive range of NLP subfields and applications:

• Core NLP Tasks: Named entity recognition, sentiment analysis, question answering, summarization, and machine translation
• Language Understanding: Discourse and pragmatics, semantics, syntax, and linguistic theories that inform cognitive modeling
• Modern AI Systems: Language modeling, large language model (LLM) agents, efficient methods for NLP, and interpretability of AI models
• Emerging Challenges: Multimodality and language grounding to vision and robotics, NLP combined with code models, and symbolic reasoning approaches
• Real-World Applications: Clinical and biomedical NLP, dialogue systems, information extraction, and computational social science
• Responsible AI: Ethics, bias, fairness, and human-centered NLP with human-AI interaction at the forefront

This breadth reflects the field's recognition that NLP is no longer confined to text analysis alone. The conference acknowledges that modern language models operate across multiple modalities, integrate with code systems, and raise fundamental questions about how humans and machines should interact.

How to Submit Your Research to EACL 2027

The conference uses a two-stage submission process through the ACL Rolling Review (ARR) system and the OpenReview platform. Here is what researchers need to know:

• Initial Submission: Authors submit papers to ARR by August 3, 2026, where they receive reviews and meta-reviews from ARR reviewers and action editors
• Commitment Phase: Authors with reviewed papers can then commit their work to EACL 2027 by October 11, 2026, where Senior Area Chairs and Program Chairs make final acceptance decisions
• Paper Formats: Long papers can be up to 8 pages of content plus unlimited references and appendices; short papers can be up to 4 pages of content plus unlimited references and appendices
• Key Deadlines: ARR submission closes August 3, 2026; reviewer registration deadline is August 5, 2026; author response period runs September 14-19, 2026; meta-reviews release October 8, 2026; and the EACL commitment deadline is October 11, 2026
• Presentation Requirement: All accepted papers must be presented at the conference in either in-person or virtual format to appear in the proceedings

An important change for this cycle separates the author response and author-reviewer discussion into two distinct stages. From September 14 to 19, 2026, authors respond to reviewers without reviewer engagement. Then from September 20 to 24, 2026, reviewers and authors engage in discussion and reviewers finalize their reviews.

The conference leadership includes General Chair Isabelle Augenstein from the University of Copenhagen, alongside Program Chairs Malvina Nissim from the University of Groningen, Roi Reichart from the Technion-Israel Institute of Technology, and Sara Tonelli from Fondazione Bruno Kessler.

What Does This Mean for the Broader NLP Landscape?

The emphasis on human-centered NLP comes at a moment when the field is grappling with fundamental questions about how language models should be built and evaluated. While generative AI in analytics and language technologies are expanding rapidly, with the market growing from $1.69 billion in 2025 to $2.21 billion in 2026, the research community is stepping back to ask whether current approaches adequately account for human language science.

The conference's special theme suggests that researchers believe the answer is no. By centering human language science, EACL 2027 is positioning itself as a venue where the field can reconnect with foundational insights from linguistics, cognitive science, and psycholinguistics. This could influence how NLP systems are designed, evaluated, and deployed in the coming years.

For researchers working on sentiment analysis, named entity recognition, text generation, and other core NLP tasks, the message is clear: understanding how humans actually use and understand language is not a luxury but a necessity for building systems that work reliably in practice.