Challenge track
Technical reports for the LM-Playschool Challenge, at least 3 pages long and up to 5 pages of content (excluding references), detailing methodology, experimental setup, training data, and results for submitted systems.
NLP/CORE Unranked
LM Playschool Challenge 2026
The LM Playschool Challenge 2026
The LM Playschool Workshop 2026 focuses on advancing language models through interactive, dialogue-based learning in collaborative, goal-oriented environments. It bridges research in game-based benchmarking, machine language acquisition through social interaction, and interactive learning signals, with an emphasis on data efficiency, social cognition, and robust agent behavior in multi-turn settings.
Key deadlines
Verified September 9, 2026Full paper
July 18, 2026
AoE
Conference timeline
Submission and decisions
Full paperKey deadline
July 18, 2026 · AoE
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Paper-only track
Original research, work-in-progress, position papers, or recently published papers on language models learning through situated interaction; can be short (up to 4 pages) or long (up to 8 pages), and may be archival or non-archival.
Research areas in scope
01
Topics of Interest
Architectures and training regimes for interactive agentsIntrinsic rewards and learning signals (RL from game-state success)Benchmarking via dialogue gamesData efficiency and social interactionSocial cognition and Theory of Mind in interactive systemsHuman-agent collaboration and coordinationEmbodied interactive agentsCommunicative and perceptual grounding
Policies worth checking twice
- All submissions must follow ARR submission guidelines and use the official ACL style template.
- Submissions must be self-contained; reviewers are not required to consult appendices or supplementary materials.
- Review versions must be anonymized unless the submission is explicitly non-archival and based on already published work.
- Accepted archival papers will appear in the workshop proceedings.
- At least one author of each accepted paper must present the work at the workshop.
- Submissions must comply with ARR policies on publication ethics, citation and comparison, multiple submission, use of writing assistance, limitations, ethical considerations, and responsible NLP research.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026