#combinatorial-reasoning News & Analysis

2 articles tagged with #combinatorial-reasoning. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.

2 articles

AINeutralarXiv – CS AI · Mar 37/104

🧠

Revealing Combinatorial Reasoning of GNNs via Graph Concept Bottleneck Layer

Researchers developed a new graph concept bottleneck layer (GCBM) that can be integrated into Graph Neural Networks to make their decision-making process more interpretable. The method treats graph concepts as 'words' and uses language models to improve understanding of how GNNs make predictions, achieving state-of-the-art performance in both classification accuracy and interpretability.

AINeutralarXiv – CS AI · Jun 196/10

🧠

CombEval: A Framework for Evaluating Combinatorial Counting in Large Language Models

Researchers introduce CombEval, a dynamic benchmark framework for evaluating how well large language models handle combinatorial counting problems. Testing 11 LLMs reveals significant brittleness in handling ordered objects, indistinguishable elements, and nested dependencies, with code-augmented approaches showing modest improvements over direct reasoning.