A Foundation Model for Zero-Shot Logical Rule Induction

Hugging Face Daily Papers Papers

Summary

This paper introduces the Neural Rule Inducer (NRI), a foundation model for zero-shot logical rule induction that uses domain-agnostic statistical properties to generalize across tasks without retraining.

Inductive Logic Programming (ILP) learns interpretable logical rules from data. Existing methods are transductive: their learned parameters are bound to specific predicates and require retraining for each new task. We introduce Neural Rule Inducer (NRI), a pretrained model for zero-shot rule induction. Rather than encoding literal identities, NRI represents literals using domain-agnostic statistical properties such as class-conditional rates, entropy, and co-occurrence, which generalize across variable identities and counts without retraining. The model consists of a statistical encoder and a parallel slot-based decoder. Parallel decoding preserves the permutation invariance of logical disjunction; an autoregressive decoder would instead impose an arbitrary clause order. Product T-norm relaxation makes rule execution differentiable, allowing end-to-end training on prediction accuracy alone. We evaluate NRI on rule recovery, robustness to label noise and spurious correlations, and zero-shot transfer to real-world benchmarks, and we believe this work opens up the possibility of foundation models for symbolic reasoning. Code and the reference checkpoint are available at https://github.com/phuayj/neural-rule-inducer.
Original Article
View Cached Full Text

Cached at: 05/08/26, 07:44 AM

Paper page - A Foundation Model for Zero-Shot Logical Rule Induction

Source: https://huggingface.co/papers/2605.04916

Abstract

Neural Rule Inducer (NRI) enables zero-shot rule induction by representing literals through domain-agnostic statistical properties and using parallel decoding to maintain permutation invariance in logical disjunctions.

Inductive Logic Programming(ILP) learns interpretable logical rules from data. Existing methods are transductive: their learned parameters are bound to specific predicates and require retraining for each new task. We introduce Neural Rule Inducer (NRI), apretrained modelforzero-shot rule induction. Rather than encoding literal identities, NRI represents literals using domain-agnostic statistical properties such as class-conditional rates, entropy, and co-occurrence, which generalize across variable identities and counts without retraining. The model consists of astatistical encoderand a parallelslot-based decoder. Parallel decoding preserves thepermutation invarianceoflogical disjunction; anautoregressive decoderwould instead impose an arbitrary clause order.Product T-norm relaxationmakes rule execution differentiable, allowing end-to-end training on prediction accuracy alone. We evaluate NRI on rule recovery, robustness to label noise and spurious correlations, and zero-shot transfer to real-world benchmarks, and we believe this work opens up the possibility of foundation models for symbolic reasoning. Code and the reference checkpoint are available at https://github.com/phuayj/neural-rule-inducer.

View arXiv pageView PDFProject pageGitHub1Add to collection

Get this paper in your agent:

hf papers read 2605\.04916

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper1

#### phuayj/neural-rule-inducer Tabular Classification• Updatedabout 17 hours ago • 68

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.04916 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.04916 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

MindZero: Learning Online Mental Reasoning With Zero Annotations

arXiv cs.AI

MindZero introduces a self-supervised reinforcement learning framework that trains multimodal large language models for efficient and robust online mental reasoning without requiring mental state annotations, outperforming model-based methods in accuracy and efficiency.