v284: Proceedings of NeSy 2025

2026-06-04 · Source: Proceedings of Machine Learning Research · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems, Emerging Technologies & Innovation · Depth: Expert, medium

Summary

Volume 284 presents the proceedings of The 19th International Conference on Neurosymbolic Learning and Reasoning, held from September 8-10, 2025, at UC Santa Cruz, CA, USA. This collection features 50 research papers exploring the integration of neural and symbolic AI paradigms. Key topics include adapting graph-based analysis for knowledge extraction from Transformer models, investigating graph neural network states, and developing scalable approaches for probabilistic neuro-symbolic robustness verification. Other contributions cover symbolic knowledge integration in reinforcement learning, learnability studies of RNNs, and generating safety-critical automotive C-programs using LLMs with formal verification. The volume also addresses enhancing large language models with neurosymbolic reasoning for multilingual tasks, explainable AI for depression detection, and frameworks for visual functional affordance grounding and commonsense reasoning in embodied agents.

Key takeaway

For AI Scientists and Research Scientists exploring advanced AI architectures, this volume highlights the critical role of neurosymbolic integration. You should consider how combining neural networks with symbolic reasoning can enhance model interpretability, improve robustness, and enable more complex reasoning in your systems. Focus on frameworks that bridge these paradigms for applications like formal verification, robot learning, or advanced LLM capabilities to push beyond current limitations.

Key insights

Neurosymbolic AI research is rapidly advancing, integrating neural networks with symbolic reasoning across diverse applications.

Principles

• Combine neural and symbolic strengths.
• Enhance AI interpretability.
• Improve system robustness via formal methods.

In practice

• Enhance LLM reasoning for complex tasks.
• Develop robust robot learning systems.
• Verify safety-critical code formally.

Topics

• Neurosymbolic AI
• Large Language Models
• Knowledge Graphs
• Formal Verification
• Explainable AI
• Reinforcement Learning

Code references

• TomPelletreauDuris/Probing-GNN-representations
• EVENFLOW-project-EU/nesy-veri
• AdritaBarua/DL-learner-using-LLMs
• HEmile/independence-vs-rs
• bradleypallen/bilateral-factuality-evaluation

Best for: AI Scientist, Research Scientist

Related on AIssential

• A neurosymbolic Approach with Epistemic Deep Learning for Hierarchical Image Classification — A neurosymbolic framework enhances hierarchical image classification with epistemic uncertainty and logical consistency.
• System 3 AI: No Humans Needed — System 3 AI integrates formal logic and verifiable code into LLMs for autonomous, complex reasoning beyond human feedback.
• Neuro-Symbolic AI for Cybersecurity: State of the Art, Challenges, and Opportunities — Neuro-Symbolic AI combines neural pattern recognition with symbolic reasoning to enhance cybersecurity defense and offense.
• Neurosymbolic Repo-level Code Localization — Neurosymbolic AI combining LLMs and Datalog overcomes keyword shortcuts for precise code localization.
• Grounding vs. Compositionality: On the Non-Complementarity of Reasoning in Neuro-Symbolic Systems — Symbol grounding alone is insufficient for compositional generalization; explicit reasoning objectives are required.
• v288: Proceedings of Neurosymbolic Systems 2025 — Neuro-symbolic AI integrates neural learning with symbolic reasoning for robust, verifiable, and interpretable intelligent systems.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Proceedings of Machine Learning Research.