Latest Neuro-Symbolic AI Research Papers

The newest Neuro-Symbolic AI papers from across the field — arXiv, NeurIPS, CVPR, Nature, and more — refreshed daily and ranked by relevance. Distill AI tracks Neuro-Symbolic AI so you don’t have to: get the standout work delivered to your inbox every morning, with 2-sentence summaries and the option to chat with any paper.

Recent papers

• Formalize, Don't Optimize: The Heuristic Trap in LLM-Generated Combinatorial Solvers
  Haoyu Wang, Yuliang Song, Tao Li, Zhiwei Deng et al. · arXiv · May 12, 2026

  Large Language Models (LLMs) struggle to solve complex combinatorial problems through direct reasoning, so recent neuro-symbolic systems increasingly use them to synthesize executable solvers. A central design question is how the LLM should…

• Abductive Reasoning with Probabilistic Commonsense
  Joseph Cotnareanu, Chiara Roverato, Han Zhou, Didier Chetelat et al. · arXiv · May 8, 2026

  Recent efforts to improve the reasoning abilities of Large Language Models (LLMs) have focused on integrating formal logic solvers within neurosymbolic frameworks. A key challenge is that formal solvers lack commonsense world knowledge, pre…

• Canonical Audit Checkpoints as Formal Falsification: Verified Boundary Testing in Neuro-Symbolic Systems
  Nick Askamp · Zenodo (CERN European Organ... · May 7, 2026

  <strong>Abstract</strong> This paper introduces a formal methodology for verified boundary testing in neuro-symbolic AI systems via <strong>canonical audit checkpoints</strong>. Instead of focusing on success rates, this framework posits th…

• Towards Neuro-symbolic Causal Rule Synthesis, Verification, and Evaluation Grounded in Legal and Safety Principles
  Zainab Rehan, Christian Medeiros Adriano, Sona Ghahremani, Holger Giese · AAMAS 2026 · Apr 30, 2026

  Rule-based systems remain central in safety-critical domains but often struggle with scalability, brittleness, and goal misspecification. These limitations can lead to reward hacking and failures in formal verification, as AI systems tend t…

• Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models
  Gongbo Zhang, Wen Wang, Ye Tian, Li Yuan · arXiv · Apr 29, 2026

  Diffusion large language models (dLLMs) offer parallel decoding and bidirectional context, but state-of-the-art dLLMs require billions of parameters for competitive performance. While existing distillation methods for dLLMs reduce inference…

• From Protocol to Practice: Graded Sepsis Bundle Compliance and Actionable Insights from Real-World ICU Data
  HIMANSHU TRIPATHI, Kaushik Roy, Shahram Rahimi, Subash Neupane et al. · medRxiv · Apr 25, 2026

  Abstract Sepsis is a leading cause of in-hospital mortality, yet systematically evaluating temporal adherence to the Surviving Sepsis Campaign (SSC) bundle across large patient populations remains difficult due to semantic variability in el…

• GRAIL: Autonomous Concept Grounding for Neuro-Symbolic Reinforcement Learning
  Hikaru Shindo, Henri Rößler, Quentin Delfosse, Kristian Kersting · arXiv (Cornell University) · Apr 18, 2026

  Neuro-symbolic Reinforcement Learning (NeSy-RL) combines symbolic reasoning with gradient-based optimization to achieve interpretable and generalizable policies. Relational concepts, such as "left of" or "close by", serve as foundational bu…

• Explainable neuro-symbolic artificial intelligence for automated interpretation of corneal topography and early keratoconus detection
  Han Wang, Shuai Qin · Frontiers in Artificial Int... · Apr 13, 2026

  Background: Early detection of keratoconus is essential for preventing postoperative complications in refractive surgery and preserving long-term visual function. Although artificial intelligence has demonstrated strong potential in ophthal…

• Towards a neuro-symbolic approach for precision anti-reflux surgery
  Quan Wang, Yaowei Dai, Alberto Aiolfi, Marco Manna et al. · Updates in Surgery · Apr 10, 2026

  Surgical management of gastroesophageal reflux disease (GERD) is limited by non-technical challenges, including variability in patient selection, incomplete physiological assessment, imprecise procedure choice, and heterogeneity of intraope…

• FedLTN-CubeSat: Neuro-Symbolic Federated Learning for Intrusion Detection in LEO CubeSat Constellations
  Gang Yang, Lin Ni, Junfeng Geng, Xiang Peng · Mathematics · Mar 20, 2026

  Low Earth Orbit (LEO) mega-constellations are becoming the backbone of global communications, yet their cybersecurity remains critically under-addressed. Intrusion detection systems (IDSs) for such constellations face a unique trilemma of a…

• ARTEMIS: A Neuro Symbolic Framework for Economically Constrained Market Dynamics
  Rahul D Ray · ArXiv.org · Mar 18, 2026

  Deep learning models in quantitative finance often operate as black boxes, lacking interpretability and failing to incorporate fundamental economic principles such as no-arbitrage constraints. This paper introduces ARTEMIS (Arbitrage-free R…

• CausalTrace: A Neurosymbolic Causal Analysis Agent for Smart Manufacturing
  Chathurangi Shyalika, Aryaman Sharma, Fadi El Kalach, Utkarshani Jaimini et al. · Proceedings of the AAAI Con... · Mar 14, 2026

  Modern manufacturing environments demand not only accurate predictions but also interpretable insights to process anomalies, root causes, and potential interventions. Existing AI systems often function as isolated black boxes, lacking the s…