Full papers
Original research contributions with comprehensive experiments and analysis.
NLP/CORE Unranked
RL4VLA 2026
Reinforcement Learning for Vision-Language-Action Models Workshop
The RL4VLA 2026 workshop focuses on reinforcement learning for vision-language agents, bringing together researchers to explore advances in training, evaluation, and applications of agents that integrate visual and linguistic understanding through reinforcement learning techniques.
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Short papers
Preliminary results, position papers, or work-in-progress with clear insights.
Demo papers
Demonstrations of systems or tools related to RL for vision-language agents.
Research areas in scope
01
Topics
Reinforcement learning for vision-language agentsMultimodal representation learningLanguage-conditioned policy learningVision-language grounding and reasoningBenchmarking and evaluation of vision-language agentsHuman-in-the-loop and interactive learningScalable training methods for vision-language RLReal-world deployment and safety of vision-language agents
Policies worth checking twice
- Submissions must be anonymized for double-blind review.
- Papers must not exceed 8 pages for full papers and 4 pages for short/demo papers, excluding references.
- Dual submission to other conferences or journals is not allowed during review.
- Author lists cannot be changed after the submission deadline.
- Authors must include an AI-use statement describing any generative AI tools used in the research.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026