Logically Consistent Language Models via Neuro-Symbolic Integration | Diego Calanzone

💬 Discord: https://discord.gg/h8NVzwnysW 💻 GitHub: https://github.com/centaurinstitute 🤝 LinkedIn: https://www.linkedin.com/company/centaur-ai-institute/ 📢 New Initiative: Neuro-Symbolic Agentic Protocol ⭐ Star us on GitHub https://github.com/centaurinstitute/neuro-symbolic-agentic-protocol — ​​📘 This week, Diego Calanzone will presents that Large Language Models (LLMs) are a promising venue for natural language understanding and generation. However, current LLMs are far from reliable: they are prone to generating non-factual information and, more crucially, to contradicting themselves when prompted to reason about relations between entities of the world. These problems are currently addressed with large scale fine-tuning or by delegating reasoning to external tools. In this work, we strive for a middle ground and introduce a loss based on neuro-symbolic reasoning that teaches an LLM to be logically consistent with an external set of facts and rules and improves self-consistency even when the LLM is fine-tuned on a limited set of facts. ​📄 Diego Calanzone's team approach also allows to easily combine multiple logical constraints at once in a principled way, delivering LLMs that are more consistent w.r.t. all constraints and improve over several baselines w.r.t. a given constraint. Moreover, the team's method allows LLMs to extrapolate to unseen but semantically similar factual knowledge, represented in unseen datasets, more systematically. ​🎓 Diego Calanzone is a Ph.D. student at Mila Quebec AI Institute and Université de Montréal under the supervision of prof. Pierre-Luc Bacon. His current work focuses on a theoretical line on temporal abstraction and high-diversity sampling in reinforcement learning, with applied projects in drug discovery and material discovery. Under the supervision of prof. Antonio Vergari and prof. Stefano Teso, Diego has conducted research on logical consistency and factuality in pre-trained large language models, motivating a novel knowledge editing technique grounded on neuro-symbolic methods. — 🗨️ ​Join us for an interactive session exploring Neuro-Symbolic AI, the emerging paradigm that blends the strengths of neural networks with symbolic reasoning. We will discuss how hybrid approaches can enhance generalization, interpretability, and reasoning, and how these methods are shaping the future of intelligent systems. Whether you’re a researcher, engineer, or simply curious about the cutting edge of AI, you’ll find an engaging space to learn, connect, and exchange ideas. Event Format: 🔍 Reading Group 🎙️ Panel Discussion 🧩 Tutorials/Workshop Series 💬 Research Roundtable 🛠️ Open-Source Project Review 📄 Paper Pitch 🤝 Collaboration Hour 🍵 Networking Mixer #AI #NeuroSymbolic #FutureOfAI