Skip to content
conftrace_

power-law decay

5 papers

Explore in graph

Co-occurring keywords · papers mentioning both

language model (2) compressible signals (1) compressible priors (1) kernel ridge regression (1) long short-term memory (1) learning curve (1) benign overfitting (1) inference efficiency (1) layer skipping (1) dynamic computation (1)

Papers

A Simple Model of Inference Scaling Laws ICML 2025 Position-Aware Depth Decay Decoding (D3): Boosting Large Language Model Inference Efficiency ACL 2025 On the Asymptotic Learning Curves of Kernel Ridge Regression under Power-law Decay NeurIPS 2023 Multi-timescale Representation Learning in LSTM Language Models ICLR 2021 Learning with Compressible Priors NeurIPS 2009