First Workshop for Research on Agent Language Models (REALM) at ACL 2025

Large language models (LLMs) have become a workhorse to drive intelligent agents to a higher level of agency, autonomy, and self-direction. As a result, we have witnessed a surge in LLM agents across many diverse environments including web navigation (WebArena and WorkArena), mobile applications (Mobile-Agent), protein discovery (ProtoAgent), cybersecurity (Cybench) and even scientific research (MLAgentBench). In addition, integrating various services and more advanced architecture including external tools (e.g. function calling, APIs), memory (e.g. MemGPT), external or private knowledge integration (RAG, and RETRO), advanced planning (Chain-of-Thought, ReAct, Reflection, self-critic, etc.) and multi-agent collaboration into agents has enabled them to accomplish even more complex goals. Despite these advancements, introducing this level of autonomy and complexity at large scale or in production, brings unprecedented challenges to build, configure, deploy, evaluate, collaborate and improve LLM agents safely and responsibly.

Our workshop aims to bring together researchers, practitioners, and thought leaders in the ACL community to discuss and align on the current landscape, key challenges, and future directions of LLM agents.

Venue

The REALM workshop will take place at the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), in Vienna, Austria. The conference will take place from July 27th to August 1st, 2025. Workshops will be held July 31st and August 1st. For more details about ACL 2025, check their website. Exact date and room information of the REALM workshop will be announced closer to the event date.

Important Dates

  • Submission Deadline: March 01, 2025
  • Notification of Acceptance: April 17, 2025
  • Camera Ready Deadline: May 16, 2025
  • Workshop Date: July 31 - August 01, 2025

All deadlines are calculated at 11:59pm Anywhere on Earth (UTC-12).

Invited Speakers

We have an exciting lineup of speakers. Stay tuned for more details!

Nicolas Chapados
ServiceNow Research
Daniel Fried
Carnegie Mellon University
Chris Manning
Stanford University
Siva Reddy
McGill University / Mila
Yu Su
Ohio State University
Tao Yu
University of Hong Kong

Questions

If you have any questions, please do not hesitate to contact us at realm-workshop@gmail.com.

Organizers

Ehsan Kamalloo
ServiceNow Research
Nicolas Gontier
ServiceNow Research
Xing Han Lu
McGill University / Mila
Shikhar Murty
Stanford University
Alexandre Lacoste
ServiceNow Research
Nouha Dziri
Ai2

Advisory Board

Hannaneh Hajishirzi

University of Washington / Ai2

Graham Neubig

Carnegie Mellon University / All Hands AI