docs(paper): extend continual-learning references for 2024 and 2025

Author: biru-codeastromer

paper/refs.bib

@@ -101,3 +101,10 @@ @book{wang2025continualrlbook
   publisher={Springer},
   year={2025}
 }
+
+@article{wang2024continualsurvey,
+  title={A Comprehensive Survey of Continual Learning: Theory, Method and Application},
+  author={Wang, Liyuan and Zhang, Xingxing and Su, Hang and Zhu, Jun},
+  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
+  year={2024}
+}

paper/worldmodelgym.tex

@@ -26,7 +26,7 @@ \section{Motivation and Research Questions}
 \end{enumerate}
 
 \section{Related Work}
-WorldModel Gym builds on model-based RL and latent planning, including World Models \citep{ha2018worldmodels}, PlaNet \citep{hafner2019planet}, Dreamer \citep{hafner2019dreamer}, and MuZero \citep{schrittwieser2020muzero}. We include classical online planning references such as POMCP \citep{silver2010pomcp} and DESPOT \citep{somani2013despot,ye2017despot}. For benchmark design, we draw from DeepMind Lab \citep{beattie2016dmlab}, Procgen \citep{cobbe2019procgen,cobbe2021ppg}, and Crafter \citep{hafner2021crafter}. Continual evaluation follows recent continual RL survey and synthesis efforts \citep{pan2025crlsurvey,wang2025continualrlbook}.
+WorldModel Gym builds on model-based RL and latent planning, including World Models \citep{ha2018worldmodels}, PlaNet \citep{hafner2019planet}, Dreamer \citep{hafner2019dreamer}, and MuZero \citep{schrittwieser2020muzero}. We include classical online planning references such as POMCP \citep{silver2010pomcp} and DESPOT \citep{somani2013despot,ye2017despot}. For benchmark design, we draw from DeepMind Lab \citep{beattie2016dmlab}, Procgen \citep{cobbe2019procgen,cobbe2021ppg}, and Crafter \citep{hafner2021crafter}. Continual evaluation follows recent continual RL and continual-learning survey efforts \citep{pan2025crlsurvey,wang2025continualrlbook,wang2024continualsurvey}.
 
 \section{Benchmark Design}
 \paragraph{Task families.}