Just-In-Time Reinforcement Learning: Continual Learning in LLM Agents Without Gradient Updates

Yibo Li, Zijie Lin, Ailin Deng, Xuan Zhang, Yufei He, Shuo Ji, Tri Cao, Bryan Hooi

BibTeX
@inproceedings{li2026just,
  author = {Yibo Li and Zijie Lin and Ailin Deng and Xuan Zhang and Yufei He and Shuo Ji and Tri Cao and Bryan Hooi},
  title = {{Just-In-Time Reinforcement Learning: Continual Learning in LLM Agents Without Gradient Updates}},
  booktitle = {International Conference on Machine Learning},
  year = {2026},
  url = {https://openreview.net/forum?id=pLvye0zHUC},
}
Figure