\begin{table}[t]
\caption{Representative studies related to noisy labels or noisy supervision, 2025.}\label{tab:recent2025}
\small
\begin{tabular}{@{}L{0.23\textwidth}L{0.15\textwidth}L{0.22\textwidth}L{0.30\textwidth}@{}}
\toprule
Paper & Venue/year & Main setting & Role in this review \\
\midrule
Open-world noisy data \citep{pan2025openworld} & AAAI 2025 & Open-world LNL & Class-independent margins for known/unknown noise \\
Multi-prototype open-set LNL \citep{zhang2025multiprototype} & Pattern Recognition 2025 & Open-set LNL & Feature-prototype modeling for clean/ID-noisy/OOD-noisy separation \\
ANNE \citep{cordeiro2025anne} and surrogate selection \citep{liang2025surrogate} & Pattern Recognition / IJCV 2025 & Sample selection & Hybrid loss-feature selection and general reliability surrogates \\
VRI \citep{sun2025vri} and Detect-and-Correct \citep{grinberg2025detectcorrect} & IJCV / arXiv 2025 & Target refinement & Probabilistic and selective correction under uncertain targets \\
SiDyP \citep{ye2025sidyp} & KDD 2025 & LLM-generated labels & LLM annotation noise and iterative target refinement \\
BeGIN \citep{kim2025begin} & KDD 2025 & Graph benchmarks & Instance-dependent graph label noise with algorithmic and LLM-based corruptions \\
Early Cutting \citep{yuan2025earlycutting} & NeurIPS 2025 & Sample selection & Filtering mislabeled easy examples after early confidence \\
Joint asymmetric loss \citep{wang2025jal} and RML++ \citep{li2025rmlpp} & ICCV / IJCV 2025 & Robust loss & Robust objective design beyond symmetric losses \\
ELDET \citep{choi2025eldet} & NeurIPS 2025 & Object detection & Early-learning distillation under localization and categorization noise \\
\botrule
\end{tabular}
\end{table}