No official code found. Best implementation: zhouzhengqd/ROME (9 stars, last commit 2025-07-01). 1 more below.
Multi-Track Message Passing: Tackling Oversmoothing and Oversquashing in Graph Learning via Preventing Heterophily Mixing
HGAP: Boosting Permutation Invariant and Permutation Equivariant in Multi-Agent Reinforcement Learning via Graph Attention Network
Understanding and Mitigating Hallucination in Large Vision-Language Models via Modular Attribution and Intervention
Understanding Few-Shot Learning: Measuring Task Relatedness and Adaptation Difficulty via Attributes
CLR-GAN: Improving GANs Stability and Quality via Consistent Latent Representation and Reconstruction
Addressing Spatial-Temporal Heterogeneity: General Mixed Time Series Analysis via Latent Continuity Recovery and Alignment
@article{zhou2026rome,
title = {ROME is Forged in Adversity: Robust Distilled Datasets via Information Bottleneck},
author = {Zheng Zhou and Wenquan Feng and Qiaosheng Zhang and Shuchang Lyu and Qi Zhao and Guangliang Cheng},
year = {2026},
url = {https://api.semanticscholar.org/CorpusID:52014b3ddf877fb14f9ad3325307184da08eebb4},
journal = {ICML 2025 2025}
}Shows whether official code is linked for this paper.
[](https://codeofpaper.com/paper/52014b3ddf877fb14f9ad3325307184da08eebb4)