Add related paper citation

README.md

@@ -34,7 +34,15 @@ Both runners produce JSON and CSV artifacts in the Space working directory. The
 
 It installs `lbw-guard` from PyPI and does not vendor the local `lbw/` source folder.
 
-Paper: https://arxiv.org/abs/2605.19008
+## Related Paper
+
+**Learn-by-Wire Training Control Governance: Bounded Autonomous Training Under Stress for Stability and Efficiency**
+
+- Author: Anis Radianis
+- arXiv: https://arxiv.org/abs/2605.19008
+- DOI: https://doi.org/10.48550/arXiv.2605.19008
+
+If you use these runners or the `lbw-guard` package in evaluation, please cite the paper above.
 
 Use GPU hardware for meaningful runtime. CPU can load the app, but training is intentionally capped to tiny smoke settings so users do not accidentally start long CPU jobs.