arXiv Machine Learning
RACE: Relation-Level Counterfactual Explanations for Heterogeneous Graph Neural Networks
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv Machine Learning
Sep 7
A Comparative Study of Counterfactual Explainers for Graph Neural Networks Enabling Multiple Types of Graph Edit
The paper presents a comparative analysis of six state‑of‑the‑art counterfactual explainers for graph neural networks, focusing on methods that can both add and remove edges to alter model predictions. It evaluates these explainers across diverse real‑world and synthetic datasets, covering binary and multi‑class graph and node classification tasks, using a range of quantitative and qualitative metrics. The study highlights the trade‑offs between explanation size, coverage, and quality, aiming to pinpoint each method’s strengths and weaknesses to inform future research.
arXiv Machine Learning
Aug 13
AutoGrable: What Is a Good Graph for a Table?
arXiv:2608. 11431v1 Announce Type: new Abstract: Graph learning presupposes a graph, and tables and relational databases do not come with one.
arXiv Computation and Language
Sep 3
HyGRAIL: Cost-Aware and Evidence-Grounded Scientific Hypothesis Discovery over Knowledge Graphs
HyGRAIL is a framework for discovering scientific hypotheses in incomplete knowledge graphs by combining a graph neural network (GNN) triage with large language model (LLM) review. The GNN scores candidate hypotheses and routes only ambiguous cases to the LLM, which receives structured evidence from the graph converted into natural language. Experiments on MatKG show HyGRAIL achieves the highest F1 score, improves over baselines, and cuts LLM calls by over 54%.
arXiv Machine Learning
Sep 11
Reification as a Transferable Vocabulary: Zero-Shot Link Prediction with Vanilla GNNs
The paper introduces a method that transforms knowledge graph facts into a fixed vocabulary representation, where each fact becomes a node linked to its subject, object, and relation type via six meta-relations. Using this representation, standard GNNs (e.g., GAT, GINE, GraphSAGE, R-GCN) trained on a single small graph can achieve zero‑shot link prediction on 40 inductive benchmarks, matching the performance of specialized foundation models like ULTRA. The approach also generalizes to relational databases, enabling foreign‑key prediction without cell values or schema text, and the authors provide code, checkpoints, and evaluation tools for all benchmarks.
arXiv Computation and Language
Sep 22
Euston: Training Away Mathematical Sycophancy Without Losing the Mathematics
Euston is an 8‑B parameter mathematical claim‑verification model that resists producing false derivations when presented with corrupted theorems. It was trained on 3,026 matched true/corrupted statement pairs generated by GraphSynth, a probabilistic factor‑graph generator, and fine‑tuned from DeepSeek‑R1‑8B using GRPO. On a balanced held‑out split, Euston’s balanced accuracy rose from 29.50 % to 63.75 %, and its discrimination gap improved from –0.5 % to +27.5 %, while maintaining comparable general mathematical ability and reducing response length and truncation rates.
Hugging Face Trending Papers
Sep 2
HyGRAIL: Cost-Aware and Evidence-Grounded Scientific Hypothesis Discovery over Knowledge Graphs
Scientific knowledge graphs organize entities and relations extracted from scientific literature, but they remain inherently incomplete. Missing typed links in such graphs can therefore represent plau...