Fitting Every Hint Is Not Enough: Premature Floods in Recurrent Algorithmic Reasoners
cs.LG, cs.AI
Submitted: 2026-10-10
Updated: 2026-10-10
Sources
- Path Independent Equilibrium Models Can Better Exploit Test-Time Computation
- Layer Normalization
- End-to-end Algorithm Synthesis with Recurrent Networks: Logical Extrapolation Without Overthinking
- On the Markov Property of Neural Algorithmic Reasoning: Analyses and Methods
- Understanding and Improving Length Generalization in Recurrent Models
- Underspecification Presents Challenges for Credibility in Modern Machine Learning
- Sigmoid-Weighted Linear Units for Neural Network Function Approximation in Reinforcement Learning
- Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
- Deep Equilibrium Algorithmic Reasoning
- Learning Graph Algorithms With Recurrent Graph Neural Networks
- A Generalist Neural Algorithmic Learner
- Adam: A Method for Stochastic Optimization
- Towards Better Out-of-Distribution Generalization of Neural Algorithmic Reasoning Tasks
- Flood and Echo Net: Algorithmically Aligned GNNs that Generalize
- SALSA-CLRS: A Sparse and Scalable Benchmark for Algorithmic Reasoning
- Latent Space Representations of Neural Algorithmic Reasoners
- From Shortcut Learning to Discrete Neural Insertion Sort
- Neural Algorithmic Reasoning Without Intermediate Supervision
- Discrete Neural Algorithmic Reasoning
- Can You Learn an Algorithm? Generalizing from Easy to Hard Problems with Recurrent Networks
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks