diff --git a/docs/paper-arabic/main.pdf b/docs/paper-arabic/main.pdf
index 989c904..37f034f 100644
Binary files a/docs/paper-arabic/main.pdf and b/docs/paper-arabic/main.pdf differ
diff --git a/docs/paper-arabic/main.tex b/docs/paper-arabic/main.tex
index 2c86b96..bcde303 100644
--- a/docs/paper-arabic/main.tex
+++ b/docs/paper-arabic/main.tex
@@ -110,7 +110,7 @@ \section{Related Work}
 \section{Data}
 \subsection{Sources}
 \begin{itemize}
-  \item \textbf{Tashkeela-full}: the complete 75M-word classical dump,
+  \item \textbf{Tashkeela-full}~\cite{tashkeela}: the complete 75M-word classical dump,
         cleaned with a Sadeed-style pipeline (orthographic normalization,
         haraqat consistency checks, length filtering).
   \item \textbf{Arabic Wikipedia}: modern prose, distilled of
@@ -323,6 +323,7 @@ \section{Reproducibility}
 \bibitem{mtp} Gloeckle et al. Better \& Faster Large Language Models via Multi-token Prediction. arXiv:2404.19737, 2024.
 \bibitem{gkd} Agarwal et al. On-Policy Distillation of Language Models: Learning from Self-Generated Mistakes. ICLR, 2024. arXiv:2306.13649.
 \bibitem{gtpo} Tan et al. GTPO and GRPO-S: Token and Sequence-Level Reward Shaping with Policy Entropy. arXiv:2508.04349, 2025.
+\bibitem{tashkeela} Taha Zerrouki and Amar Balla. Tashkeela: Novel corpus of Arabic vocalized texts, data for auto-diacritization systems. \emph{Data in Brief}, 11:147--151, 2017.
 \end{thebibliography}
 
 \end{document}
diff --git a/docs/paper-umbrella/main.pdf b/docs/paper-umbrella/main.pdf
index ec024c8..925850e 100644
Binary files a/docs/paper-umbrella/main.pdf and b/docs/paper-umbrella/main.pdf differ
diff --git a/docs/paper-umbrella/main.tex b/docs/paper-umbrella/main.tex
index 6a4dcc3..80530c9 100644
--- a/docs/paper-umbrella/main.tex
+++ b/docs/paper-umbrella/main.tex
@@ -72,9 +72,9 @@ \section{System and results}
 Arabic & diacritization & \textbf{2.29/1.33\% DER} & Sadeed~\cite{sadeed} 7.29/5.26\%, Gemini 3.19/2.38\%, GLM-5.3-Flash 8.80 (same benchmark+evaluator) \\
 Hebrew & diacritization & \textbf{16.43\% DER} (Biblical) & DictaBERT~\cite{dictabert} 35.6\% (same test) \\
 Thai & G2P & \textbf{2.32\% PER} & baseline 6.37\% (same test) \\
-Persian & G2P/HA & \textbf{89.5\%} / 77.3\% (SB) & Homo-GE2PE 76.9\% \\
+Persian & G2P/HA & \textbf{89.5\%} / 77.3\% (SB) & Homo-GE2PE~\cite{homoge2pe} 76.9\% \\
 Persian & diacritization & \textbf{0.52\% CER} & (none existed) \\
-Urdu & G2P & \textbf{14.77\% CER} & epitran 60.0\% \\
+Urdu & G2P & \textbf{14.77\% CER} & epitran~\cite{epitran} 60.0\% \\
 Urdu & diacritization & \textbf{3.74\% CER} & (none existed) \\
 \bottomrule
 \end{tabular}
@@ -156,6 +156,10 @@ \section{Reproducibility}
 \bibitem{byt5} Xue et al. ByT5: Towards a token-free future with
   pre-trained byte-to-byte models. TACL, 2022. arXiv:2105.13626.
 \bibitem{dictabert} DictaBERT-large-char-menaked. Dicta, 2023--2024.
+\bibitem{homoge2pe} Qharabagh et al. Rethinking G2P with Rich Data and
+  Rule-Based Models. arXiv:2505.12973, 2025.
+\bibitem{epitran} David R. Mortensen, Siddharth Dalmia, and Patrick
+  Littell. Epitran: Precision G2P for Many Languages. \emph{LREC}, 2018.
 \end{thebibliography}
 
 \end{document}