The paper presents a systematic comparison of importance‑attribution methods for deep learning models that operate on relational databases. It evaluates four methods—Monte Carlo Shapley, Kernel SHAP, single‑attribute impact, and mask learning—across regression and classification tasks using GraphSAGE and Relational Graph Transformers. The study finds that method choice affects attribute rankings and score distributions, with single‑attribute impact often matching Shapley‑based methods at lower cost, especially in regression settings.
By Hila Levi-Yosefi, Shunit Agmon, Ester Livshits, Agapi Rissaki, Benny Kimelfeld
arXiv:2606. 11198v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) systems inject external knowledge to improve LLM outputs, yet the format of injected content -- distinct from its semantic relevance -- can independently distort the model's attention distribution.
By Yuqi Zhang, Di Zhang
arXiv:2512. 12477v2 Announce Type: replace Abstract: Estimating node importance in heterogeneous knowledge graphs is a fundamental problem underlying recommendation, search, and knowledge decision systems.
By Jiawen Chen, Yanyan He, Qi Shao, Mengli Wei, Duxin Chen, Wenwu Yu, Yanlong Zhao
arXiv:2512. 12477v3 Announce Type: replace Abstract: Estimating node importance in heterogeneous knowledge graphs is a fundamental problem underlying recommendation, search, and knowledge decision systems.
By Jiawen Chen, Yanyan He, Qi Shao, Mengli Wei, Duxin Chen, Wenwu Yu, Yanlong Zhao
The paper examines a multimodal approach that combines a self‑supervised GNN encoder with an alternating optimization scheme involving a language‑model teacher. Despite the expectation that this joint strategy would enhance predictive performance, the authors find that the combined model fails to deliver significant gains. They identify six key factors—ranging from anchor strength trade‑offs to misaligned representation spaces—that explain why the integration of text knowledge does not fully benefit graph learning.
By Fumiaki Kimino (SOKENDAI), Ryoma Sato (SOKENDAI, National Institute of Informatics)
arXiv:2606. 05756v1 Announce Type: new Abstract: Graph Neural Networks (GNNs) have demonstrated remarkable performance across a range of applications involving graph-structured data, particularly in high-stakes domains.
By Jialiang Yin, Zheng Zhao, Linsey Pang, Bo Dong, Bin Shi, Jiaxing Zhang
arXiv:2607. 19042v1 Announce Type: cross Abstract: Neural hypergraphs are a natural generalization of neural networks, the reference models in modern machine learning.
By Gianluca Peri, Diego Febbe, Duccio Fanelli
The paper evaluates ten deep learning optimizers across models ranging from 0.1B to 1.2B parameters and varying data-to-model ratios, highlighting that fair comparisons require rigorous hyperparameter tuning and end-of-training evaluations. It finds that claimed speedups over AdamW are lower than advertised, dropping to only 1.1× for the largest models, and that intermediate checkpoint comparisons can mislead due to learning‑rate decay. The study shows that matrix‑based optimizers like Muon and Soap achieve the highest speedups, but these benefits diminish as model size increases.
By Kaiyue Wen, David Hall, Tengyu Ma, Percy Liang
The paper introduces Nexus, an optimizer that encourages gradient similarity during pretraining to bring task‑specific minima closer together. Experiments on models from 130M to 3B parameters show that Nexus achieves the same pretraining loss as standard optimizers while significantly improving downstream performance, including a 15% accuracy boost on GSM8k and a 0.012 reduction in out‑of‑distribution loss for the 3B model.
By Huanran Chen, Huaqing Zhang, Xiao Li, Yinpeng Dong, Ke Shen, Jun Zhu
arXiv:2603. 14894v3 Announce Type: replace-cross Abstract: Trust and ethical concerns due to the widespread deployment of opaque machine learning (ML) models motivating the need for reliable model explanations.
By Sumedha Chugh, Ranjitha Prasad, Nazreen Shah
The paper introduces Ra‑NEM, a method that optimizes the area under insertion and deletion curves to improve the faithfulness of feature importance attributions in machine learning models. By deriving an objective function from faithfulness and approximating its gradient, the authors link insertion curves to top‑k feature selection, enabling a loss function that measures attribution quality. Ra‑NEM, integrated with the neural explanation mask framework, works with any differentiable model, yielding accurate, robust, and fast attributions that outperform existing algorithms on multiple XAI metrics.
By Bj{\o}rn Leth M{\o}ller, Bulat Ibragimov, Christian Igel
HyperFuse is a new, label‑free pipeline for fast self‑supervised node embeddings on attributed hypergraphs. It computes structural node coordinates via a spectral relaxation of hypergraph modularity, builds multi‑scale feature summaries with utility‑weighted hyperedges, and trains a lightweight encoder for only 100 epochs. In experiments on nine public hypergraphs, HyperFuse achieved 13–179× speed‑ups over baselines and matched or exceeded their accuracy on most downstream tasks.
By Megha P, Harshit Kumar, Srajan Agarwal, Anirban Banerjee, Olaf Wolkenhauer, Saptarshi Bej