Structural Generalization

Follow this topic Ask about this topic

Recent momentum

+33%

4 papers in the last 28 days · 0.1% of indexed attention

Twelve weeks of publication activity for this topic as it is defined today.

41 papers

Latest in Structural Generalization

Open your feed →

Quantifying the Memorization-to-Generalization Transition: Scaling Laws and Phase Structure in Grokking
Sep 12, 2026Anish KatariaGrokking Over-Parameterized Neural Networks
A Multi-Stage Rule-Chaining Framework for Compositional and Interpretable Cognitive Reasoning
Sep 12, 2026Deblina KarTheory-Of-Mind Reasoning Structural Generalization
Efficient Leakage-Free Neural Architecture Search under Leave-One-Subject-Out Evaluation
Sep 10, 2026Heinke HihnNeural Architecture Search Search Algorithms
Length Generalization for Transformers via Compression
Sep 8, 2026Georg Zetzsche, Hongjian Jiang, Andy Yang +4Transformer Architectures Generalization Bounds
Language Models Generalize to Human-like Word Order Preferences
Aug 5, 2026Amanda Popadich, Shane Steinert-ThrelkeldLinguistics Structural Generalization
On the Computational Complexity of Structural Generalization
Jul 21, 2026Zichao WeiStructural Generalization Compositionality
What General Intelligence Requires: Non-Reducible Constraints Across Levels of Description
Jul 21, 2026Subhomoy BakshiArtificial Intelligence Systems Cognitive Science
Similarity All The Way Up: Multilingual Generalization in LLMs Relies on Language-Level Similarity Structures
Jul 18, 2026Supantho Rakshit, Adele Goldberg, Henry ConklinMultilingual Large Language Models Structural Generalization
What Makes a Representational Prior Work? Feature Families, Label-Free Invariances, and Critical Windows in Grokking
Jul 14, 2026Gunner Levi HowePrior Knowledge Structural Generalization
Traj-VLN: Learning Pixel-Space Interaction via Autoregressive Trajectory Generation
Jul 12, 2026Changfei Fu, Guangcheng Chen, Wenjun Xu +1Vision-Language Navigation Trajectory Prediction
Structure-Specific Representational Priors Causally Control the Grokking Delay
Jul 5, 2026Gunner Levi HoweStructural Generalization Representation Learning
Demonstrating Generalization Failures via Mixtures of Conditional Policies
Jul 3, 2026Jou Barzdukas, Jack Peck, Julian Schulz +3Structural Generalization Mixtures
On the Role of Directionality in Structural Generalization
Jul 2, 2026Zichao WeiStructural Generalization Logit Lens
The Generalization Spectrum: A Chromatographic Approach to Evaluating Learning Algorithms
Jun 24, 2026Jinghan Zhang, Zerui Cheng, Shiqi Chen +5Structural Generalization Memorization
CoRDE: Concept-Prior Routed Diffusion Experts for Structural Generalization in Robot Manipulation
Jun 20, 2026Haidong Huang, Xixin Zhao, Yaohua Zhou +5Diffusion Policies Iterative Co-Training
Formalizing Task-Space Complexity for Zero-Shot Generalization
Jun 18, 2026Jung-Hoon Cho, Heling Zhang, Siqi Du +2Structural Generalization State Space
How Post-Training Shapes Biological Reasoning Models
Jun 15, 2026Lukas Fesser, Hanlin Zhang, Michelle M. Li +5Biological Large Reasoning Models
A Compositional Framework for Open-ended Intelligence
Jun 13, 2026Ida Momennejad, Roberta RaileanuIntelligence Structural Generalization
Exploring the Scale and Diversity of Speech Anti-spoofing Datasets: Experiments and Analysis
Jun 6, 2026Zhuolin Yi, Jun Xue, Yanzhen Ren +5Cross-Dataset Benchmark Neural Scaling Laws
Mitosis Detection in the Wild: Multi-Tumor and Context-Aware Generalization in the MIDOG 2025 Challenge
Jun 5, 2026Marc Aubreville, Jonas Ammeling, Sweta Banerjee +64Computational Pathology Structural Generalization
HORIZON: Recoverability-Governed Curriculum for Physical-Domain Scaling
Jun 3, 2026Chenhao Bai, Liqin Lu, Kaijun Wang +5Robot Policies Curriculum
World-Task Factorization for Robot Learning
Jun 1, 2026Eduardo Sebastián, Adrian Pfisterer, Vito Mengers +2Robotics World Models
An In-Vitro Study on Cross-Lingual Generalization in Language Models
May 26, 2026Adrian CosmaCross-Lingual Transfer Structural Generalization
Understanding Goal Generalisation in Sequential Reinforcement Learning
May 22, 2026Jason Ross Brown, Edward James YoungStructural Generalization
A mathematical theory of balancing relational generalization and memorization
May 21, 2026Luke Cheng, Samuel LipplStructural Generalization Memorization
A Measure-Theoretic Analysis of Reasoning: Structural Generalization and Approximation Limits
May 19, 2026Yuyang Zhang, Yifu Zhang, Xuehai Zhou +1Structural Generalization Large Language Model Reasoning
Shallow ReLU $^s$ Networks in $L^p$ -Type and Sobolev Spaces: Approximation and Path-Norm Controlled GeneralizationShallow ReLU^s Networks in L^p-Type and Sobolev Spaces: Approximation and Path-Norm Controlled Generalization
May 18, 2026Weizhao Li, Fanghui Liu, Lei ShiLogarithmic Sobolev Inequalities Rectified Linear Unit Networks
Beyond Sentiment Classification: A Generative Framework for Emotion Intensity Evaluation in Text
May 15, 2026Francesco A. Fabozzi, Dasol Kim, William N. GoetzmannSentiment Analysis Structural Generalization
Differentiable Learning of Lifted Action Schemas for Classical Planning
May 13, 2026Jonas Reiter, Jakob Elias Gebler, Hector GeffnerClassical Planning Planning Methods
Understanding diffusion models requires rethinking (again) generalization
May 7, 2026Pierre Marion, Yu-Han WuOverfitting Discrete Diffusion Model
Beyond the Training Distribution: Mapping Generalization Boundaries in Neural Program Synthesis
Apr 30, 2026Henrik Voigt, Michael Habeck, Joachim GiesenStructural Generalization Mock-Interface Synthesis
Structural Generalization on SLOG without Hand-Written Rules
Apr 28, 2026Zichao WeiStructural Generalization Handwriting
Generalization and Membership Inference Attack a Practical Perspective
Apr 21, 2026Fateme Rahmani, Mahdi Jafari Siavoshani, Mohammad Hossein RohbanMembership Inference Attacks Structural Generalization
On the Emergence of Syntax by Means of Local Interaction
Apr 20, 2026Zichao WeiSyntactic Structure Context-Free Grammars
On the Generalization Bounds of Symbolic Regression with Genetic Programming
Apr 19, 2026Masahiro Nomura, Ryoki Hamano, Isao OnoSymbolic Regression Generalization Bounds
ManipArena: Comprehensive Real-world Evaluation of Reasoning-Oriented Generalist Robot Manipulation
Mar 30, 2026Yu Sun, Meng Cao, Yang Ping +24Robot Systems Manipulation
Compositional Generalization Requires Linear, Orthogonal Representations in Vision Embedding Models
Feb 27, 2026Arnas Uselis, Andrea Dittadi, Seong Joon OhStructural Generalization Implicit Neural Representations
A Theory of How Pretraining Shapes Inductive Bias in Fine-Tuning
Feb 23, 2026Nicolas Anguita, Francesco Locatello, Andrew M. Saxe +4Finetuning Feature Learning
Semi-supervised CAPP Transformer Learning via Pseudo-labeling
Feb 1, 2026Dennis Gross, Helge Spieker, Arnaud Gotlieb +3Semi-Supervised Learning Hybrid Transformer
How Causal Abstraction Underpins Computational Explanation
Aug 15, 2025Atticus Geiger, Jacqueline Harding, Thomas IcardStructural Generalization Representational Capacity
A Theoretical Analysis of Provable Compositional Generalization in Neural Networks: A Necessary and Sufficient Condition
May 5, 2025Yuanpeng LiStructural Generalization Neural Network

Sep 12, 2026 · cs.AIJ/K move · Enter open · S save

Quantifying the Memorization-to-Generalization Transition: Scaling Laws and Phase Structure in Grokking

Anish Kataria

Neural networks trained past memorization frequently undergo a delayed transition to generalization, a phenomenon known as grokking. Despite theoretical progress on \emph{why} this transition occurs, the quantitative structure of \emph{when} it occurs in hyperparameter space remains uncharacterized. We map the memorization-to-generalization boundary across 384 configurations of two-hidden-layer MLPs on modular arithmetic, fitting a power-law scaling relation for generalization onset time:

T_{\mathrm{grok}} \propto H^{-0.27}\, D^{-2.04}\, \eta^{-0.50}\, \lambda^{-0.64}

(

R^2 = 0.732

;

0.821

with interactions). The exponent hierarchy reveals that data complexity (

D^{-2.04}

) is the dominant driver of regime transition, not model capacity (

H^{-0.27}

): doubling data accelerates generalization by

{\sim}4\times

, while doubling width yields only

{\sim}1.2\times

. A sharp phase boundary at weight decay

\lambda \gtrsim 1.0

separates grokking from non-grokking configurations, and weight norm trajectories show monotonic compression during the transition, consistent with implicit regularization selecting low-complexity solutions. These results provide a quantitative foundation for predicting and controlling regime transitions in overparameterized networks.

Grokking Over-Parameterized Neural Networks Two-Layer Neural Networks Memorization Scaling Laws Structural Generalization Exponents Transitions

Open paper Ask about it PDF