Cover Image

Shortcut Solutions Learned by Transformers Impair Continual Compositional Reasoning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Redman, William T., Johnson, Erik C., Robinson, Brian
Format:	Preprint
Published:	2026
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Reasoning in Transformers -- Mitigating Spurious Correlations and Reasoning Shortcuts
by: Enström, Daniel, et al.
Published: (2024)

ShortcutProbe: Probing Prediction Shortcuts for Learning Robust Models
by: Zheng, Guangtao, et al.
Published: (2025)

Neurosymbolic Reasoning Shortcuts under the Independence Assumption
by: van Krieken, Emile, et al.
Published: (2025)

Koopman Learning with Episodic Memory
by: Redman, William T., et al.
Published: (2023)

MiMu: Mitigating Multiple Shortcut Learning Behavior of Transformers
by: Zhao, Lili, et al.
Published: (2025)

Generative Classifiers Avoid Shortcut Solutions
by: Li, Alexander C., et al.
Published: (2025)

On the Foundations of Shortcut Learning
by: Hermann, Katherine L., et al.
Published: (2023)

Efficient Unsupervised Shortcut Learning Detection and Mitigation in Transformers
by: Kuhn, Lukas, et al.
Published: (2025)

MMLU-Pro+: Evaluating Higher-Order Reasoning and Shortcut Learning in LLMs
by: Taghanaki, Saeid Asgari, et al.
Published: (2024)

Right for the Right Reasons: Avoiding Reasoning Shortcuts via Prototypical Neurosymbolic AI
by: Andolfi, Luca, et al.
Published: (2025)

Be Persistent: Towards a Unified Solution for Mitigating Shortcuts in Deep Learning
by: Dolatabadi, Hadi M., et al.
Published: (2024)

On How Iterative Magnitude Pruning Discovers Local Receptive Fields in Fully Connected Neural Networks
by: Redman, William T., et al.
Published: (2024)

Models Know Their Shortcuts: Deployment-Time Shortcut Mitigation
by: Li, Jiayi, et al.
Published: (2026)

SLAP: Shortcut Learning for Abstract Planning
by: Liu, Y. Isabel, et al.
Published: (2025)

Do LLMs Overcome Shortcut Learning? An Evaluation of Shortcut Challenges in Large Language Models
by: Yuan, Yu, et al.
Published: (2024)

BEARS Make Neuro-Symbolic Models Aware of their Reasoning Shortcuts
by: Marconato, Emanuele, et al.
Published: (2024)

Mitigating Shortcut Learning with InterpoLated Learning
by: Korakakis, Michalis, et al.
Published: (2025)

Diagnosing Shortcut-Induced Rigidity in Continual Learning: The Einstellung Rigidity Index (ERI)
by: Gu, Kai, et al.
Published: (2025)

Navigating the Shortcut Maze: A Comprehensive Analysis of Shortcut Learning in Text Classification by Language Models
by: Zhou, Yuqing, et al.
Published: (2024)

Learning Realistic Joint Space Boundaries for Range of Motion Analysis of Healthy and Impaired Human Arms
by: Keyvanian, Shafagh, et al.
Published: (2023)

A Neuro-Symbolic Benchmark Suite for Concept Quality and Reasoning Shortcuts
by: Bortolotti, Samuele, et al.
Published: (2024)

From Shortcut to Induction Head: How Data Diversity Shapes Algorithm Selection in Transformers
by: Kawata, Ryotaro, et al.
Published: (2025)

Stabilizing and Solving Unique Continuation Problems by Parameterizing Data and Learning Finite Element Solution Operators
by: Burman, Erik, et al.
Published: (2024)

Short-circuiting Shortcuts: Mechanistic Investigation of Shortcuts in Text Classification
by: Eshuijs, Leon, et al.
Published: (2025)

Learning with Logical Constraints but without Shortcut Satisfaction
by: Li, Zenan, et al.
Published: (2024)

Symbol Grounding in Neuro-Symbolic AI: A Gentle Introduction to Reasoning Shortcuts
by: Marconato, Emanuele, et al.
Published: (2025)

Layer Specialization Underlying Compositional Reasoning in Transformers
by: Liu, Jing
Published: (2025)

Shortcut Learning Susceptibility in Vision Classifiers
by: Suhail, Pirzada, et al.
Published: (2025)

Rectifying Shortcut Behaviors in Preference-based Reward Learning
by: Ye, Wenqian, et al.
Published: (2025)

Neural Flow Samplers with Shortcut Models
by: Chen, Wuhao, et al.
Published: (2025)

On Measuring Localization of Shortcuts in Deep Networks
by: Tsoy, Nikita, et al.
Published: (2025)

Reasoning Stabilization Point: A Training-Time Signal for Stable Evidence and Shortcut Reliance
by: Dhayalkar, Sahil Rajesh
Published: (2026)

The Paradox of Outcome Optimization: A Causal Information-Theoretic Bound on Reasoning Shortcuts in LLMs
by: Chen, Zihan, et al.
Published: (2026)

Initialization is Critical to Whether Transformers Fit Composite Functions by Reasoning or Memorizing
by: Zhang, Zhongwang, et al.
Published: (2024)

Trust Me, I Know the Way: Predictive Uncertainty in the Presence of Shortcut Learning
by: Wimmer, Lisa, et al.
Published: (2025)

Conditional Information Bottleneck for Multimodal Fusion: Overcoming Shortcut Learning in Sarcasm Detection
by: Wang, Yihua, et al.
Published: (2025)

Learning Solutions of Stochastic Optimization Problems with Bayesian Neural Networks
by: Lahoud, Alan A., et al.
Published: (2024)

Task-Agnostic Experts Composition for Continual Learning
by: Quarantiello, Luigi, et al.
Published: (2025)

Fairness is Not Flat: Geometric Phase Transitions Against Shortcut Learning
by: Rodriguez-Alvarez, Nicolas, et al.
Published: (2026)

Inductive Subgraphs as Shortcuts: Causal Disentanglement for Heterophilic Graph Learning
by: Wang, Xiangmeng, et al.
Published: (2026)