Long papers (archival)
Original unpublished work with concrete evaluation and analysis; accepted papers appear in the ACL Anthology workshop proceedings.
NLP/CORE Unranked
REALM 2026
Second Workshop for Research on Agent Language Models (ARR Submission)
REALM 2026 is the second workshop focused on research at the intersection of AI agents and large language models, following its successful debut at ACL 2025. It aims to bring together the ACL community to analyze current challenges and future directions in building reliable, safe, and effective agent systems, particularly in areas like planning, evaluation, interaction, and security.
Key deadlines
Verified September 9, 2026Full paper
September 1, 2026
AoE
Workshop timeline
Submission and decisions
Full paperKey deadline
September 1, 2026 · AoE
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Short papers (archival)
Original unpublished work including a small focused contribution, negative result, opinion piece, or interesting application; accepted papers appear in the ACL Anthology workshop proceedings.
Non-archival papers
Previously published work, work under review elsewhere, or preliminary results; presented at the workshop but not in the proceedings; must follow same page limits as archival submissions.
Research areas in scope
01
Autonomous Planning and Decision-Making
Mechanisms and models that empower LLM agents to make plans and independent decisions in dynamic environments
02
Agent Quality Evaluation
Methodologies to assess the agent performance on complex and long-horizon tasks
03
Data and simulation environments
Data architectures, datasets and simulations for building LLM agents
04
Coding Agents
Specialized architectures, training strategies, and datasets for software engineering tasks such as code generation, debugging, refactoring, and unit testing
05
Agent Architectures
Design and implementation of architectural frameworks that support agency in LLMs, enabling flexibility, scalability, and robustness in various tasks
06
Human-Agent and Agent-Agent Interaction
Studies on how LLM agents interact with humans and other agents, including collaboration, feedback, learning, trust, and user experienceMulti-agent systems
07
Security, Safety and Robustness
Ensuring the reliability and safety of agentic LLM systems against adversarial scenarios and malicious use
08
Ethics and Governance
Frameworks and guidelines for the ethical development, deployment, and oversight of agentic LLM systems
Policies worth checking twice
- All submissions must be anonymous for double-blind review; author names and affiliations must be omitted.
- Archival submissions must not be under review at any other venue during the REALM review period.
- Non-archival submissions may overlap with previously published or concurrently submitted work.
- Submissions must follow the ACL 2026 style guidelines and be in PDF format.
- Authors must attest to ethical considerations in their work in line with the ACL Code of Ethics.
- Preprints (e.g., on arXiv) may be posted publicly at any time as long as the submitted PDF does not link to de-anonymized author information.
- If an archival paper is accepted, authors must withdraw it from any other venue where it remains under consideration.
- Each submission must nominate one author as a reviewer to help ensure at least three reviews.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026