Research map / Probabilistic and scientific modeling / Variational and generative sampling
Embeddings as Probabilistic Equivalence in Logic Programs
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a probabilistic-equivalence view of embeddings in logic programs, addressing the gap between unification and its differentiable relaxation in neurosymbolic frameworks.
Problem
Differentiable relaxations of unification used in neurosymbolic logic programs with embedding models do not behave like true unification, since the relaxed counterpart differs from the symbolic algorithm.
Approach
Reinterprets embedding similarity in logic programs as probabilistic equivalence between symbols, so that the relaxed unification is grounded in a probabilistic semantics rather than an ad hoc relaxation.
Experimental setup
Builds on: differentiable unification neural theorem provers embedding-based logic programming
Closest papers by meaning
- Interchangeable Token Embeddings for Extendable Vocabulary and Alpha-Equivalence (ICML 2025 · similarity 0.89)
- An Evaluation of Approaches to Train Embeddings for Logical Inference (Student Abstract) (AAAI 2025 · similarity 0.88)
- Improving Soft Unification with Knowledge Graph Embedding Methods (ICML 2025 · similarity 0.86)
- Right for the Right Reasons: Avoiding Reasoning Shortcuts via Prototypical Neurosymbolic AI (NeurIPS 2025 · similarity 0.86)
- Divide and Translate: Compositional First-Order Logic Translation and Verification for Complex Logical Reasoning (ICLR 2025 · similarity 0.85)
- On the Hardness of Probabilistic Neurosymbolic Learning (ICML 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.