Neural networks exhibit grokking on small algorithmic datasets, achieving perfect generalization well after overfitting.
International Conference on Learning Representations (ICLR) , year =
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
The authors derive a Maximally Scale-Stable Parameterization (MSSP) for MoE models that achieves robust learning-rate transfer and monotonic performance gains with scale across co-scaling regimes of width, experts, and sparsity.
Language models show good calibration when asked to estimate the probability that their own answers are correct, with performance improving as models get larger.
Repeating 0.1% of training data 100 times degrades an 800M parameter model's performance to that of a 400M model by damaging copying mechanisms and induction heads associated with generalization.
Ranked preference modeling outperforms imitation learning for language model alignment and scales more favorably with model size.
Effective data transferred from pre-training to fine-tuning is described by a power law in model parameter count and fine-tuning dataset size, acting like a multiplier on the fine-tuning data.
A sparse-activation model predicts double-descent loss with distinct under- and over-parameterized scaling exponents set by sparsity, plus a compute-optimal frontier favoring dataset growth.
Presents a single functional form for neural scaling that unifies multiple scaling dimensions and claims higher extrapolation accuracy than prior forms across diverse tasks and architectures.
Machine learning research should prioritize ideas by testing their predicted behavioral signatures in modern models through custom experiments instead of leaderboard chasing or abstract theorems.
This perspective article develops a definition of foundational MLIPs and poses six open questions that the authors believe will define future research in machine-learned interatomic potentials.
citing papers explorer
-
Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets
Neural networks exhibit grokking on small algorithmic datasets, achieving perfect generalization well after overfitting.
-
How to Scale Mixture-of-Experts: From muP to the Maximally Scale-Stable Parameterization
The authors derive a Maximally Scale-Stable Parameterization (MSSP) for MoE models that achieves robust learning-rate transfer and monotonic performance gains with scale across co-scaling regimes of width, experts, and sparsity.
-
Language Models (Mostly) Know What They Know
Language models show good calibration when asked to estimate the probability that their own answers are correct, with performance improving as models get larger.
-
Scaling Laws and Interpretability of Learning from Repeated Data
Repeating 0.1% of training data 100 times degrades an 800M parameter model's performance to that of a 400M model by damaging copying mechanisms and induction heads associated with generalization.
-
A General Language Assistant as a Laboratory for Alignment
Ranked preference modeling outperforms imitation learning for language model alignment and scales more favorably with model size.
-
Scaling Laws for Transfer
Effective data transferred from pre-training to fine-tuning is described by a power law in model parameter count and fine-tuning dataset size, acting like a multiplier on the fine-tuning data.
-
Asymmetric Scaling Laws from Sparse Features
A sparse-activation model predicts double-descent loss with distinct under- and over-parameterized scaling exponents set by sparsity, plus a compute-optimal frontier favoring dataset growth.
-
Unified Neural Scaling Laws
Presents a single functional form for neural scaling that unifies multiple scaling dimensions and claims higher extrapolation accuracy than prior forms across diverse tasks and architectures.
-
Position: Ideas Should be the Center of Machine Learning Research
Machine learning research should prioritize ideas by testing their predicted behavioral signatures in modern models through custom experiments instead of leaderboard chasing or abstract theorems.
-
Six Open Questions in Machine-Learned Interatomic Potential Foundation Models
This perspective article develops a definition of foundational MLIPs and poses six open questions that the authors believe will define future research in machine-learned interatomic potentials.