Cover Image

On the Unreasonable Effectiveness of Last-layer Retraining

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Hill, John C., LaBonte, Tyler, Zhang, Xinchen, Muthukumar, Vidya
Format:	Preprint
Published:	2025
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

The Group Robustness is in the Details: Revisiting Finetuning under Spurious Correlations
by: LaBonte, Tyler, et al.
Published: (2024)

Task Shift: From Classification to Regression in Overparameterized Linear Models
by: LaBonte, Tyler, et al.
Published: (2025)

General Loss Functions Lead to (Approximate) Interpolation in High Dimensions
by: Lai, Kuo-Wei, et al.
Published: (2023)

Sharp analysis of out-of-distribution error for "importance-weighted" estimators in the overparameterized regime
by: Lai, Kuo-Wei, et al.
Published: (2024)

The Unreasonable Effectiveness of LLMs for Query Optimization
by: Akioyamen, Peter, et al.
Published: (2024)

Last-iterate Convergence for Symmetric, General-sum, $2 \times 2$ Games Under The Exponential Weights Dynamic
by: Wang, Guanghui, et al.
Published: (2025)

A general technique for approximating high-dimensional empirical kernel matrices
by: Kaushik, Chiraag, et al.
Published: (2025)

Precise asymptotics of reweighted least-squares algorithms for linear diagonal networks
by: Kaushik, Chiraag, et al.
Published: (2024)

The Unreasonable Effectiveness of Entropy Minimization in LLM Reasoning
by: Agarwal, Shivam, et al.
Published: (2025)

The Unreasonable Effectiveness of Data for Recommender Systems
by: Abdou, Youssef
Published: (2026)

Thumb on the Scale: Optimal Loss Weighting in Last Layer Retraining
by: Stromberg, Nathan, et al.
Published: (2025)

The Unreasonable Effectiveness of Solving Inverse Problems with Neural Networks
by: Holl, Philipp, et al.
Published: (2024)

The Unreasonable Effectiveness of Eccentric Automatic Prompts
by: Battle, Rick, et al.
Published: (2024)

Improved and Oracle-Efficient Online $\ell_1$-Multicalibration
by: Ghuge, Rohan, et al.
Published: (2025)

The Unreasonable Effectiveness of Randomized Representations in Online Continual Graph Learning
by: Donghi, Giovanni, et al.
Published: (2025)

Beyond Semantics: The Unreasonable Effectiveness of Reasonless Intermediate Tokens
by: Valmeekam, Karthik, et al.
Published: (2025)

Estimating stationary mass, frequency by frequency
by: Nakul, Milind, et al.
Published: (2025)

Just Wing It: Near-Optimal Estimation of Missing Mass in a Markovian Sequence
by: Pananjady, Ashwin, et al.
Published: (2024)

The Unreasonable Effectiveness of Greedy Algorithms in Multi-Armed Bandit with Many Arms
by: Bayati, Mohsen, et al.
Published: (2020)

Representation Gap: Explaining the Unreasonable Effectiveness of Neural Networks from a Geometric Perspective
by: Perera, David, et al.
Published: (2026)

The Unreasonable Effectiveness of Easy Training Data for Hard Tasks
by: Hase, Peter, et al.
Published: (2024)

The Unreasonable Effectiveness Of Early Discarding After One Epoch In Neural Network Hyperparameter Optimization
by: Egele, Romain, et al.
Published: (2024)

The good, the bad and the ugly sides of data augmentation: An implicit spectral regularization perspective
by: Lin, Chi-Heng, et al.
Published: (2022)

Faster Margin Maximization Rates for Generic and Adversarially Robust Optimization Methods
by: Wang, Guanghui, et al.
Published: (2023)

The Unreasonable Effectiveness of Open Science in AI: A Replication Study
by: Gundersen, Odd Erik, et al.
Published: (2024)

Theoretical Guarantees of Data Augmented Last Layer Retraining Methods
by: Welfert, Monica, et al.
Published: (2024)

How Does the ReLU Activation Affect the Implicit Bias of Gradient Descent on High-dimensional Neural Network Regression?
by: Lai, Kuo-Wei, et al.
Published: (2026)

The Unreasonable Effectiveness of Random Target Embeddings for Continuous-Output Neural Machine Translation
by: Tokarchuk, Evgeniia, et al.
Published: (2023)

The Unreasonable Ineffectiveness of the Deeper Layers
by: Gromov, Andrey, et al.
Published: (2024)

The Unreasonable Effectiveness of Discrete-Time Gaussian Process Mixtures for Robot Policy Learning
by: von Hartz, Jan Ole, et al.
Published: (2025)

The Unreasonable Effectiveness of Model Merging for Cross-Lingual Transfer in LLMs
by: Bandarkar, Lucas, et al.
Published: (2025)

Phi-4-reasoning-vision-15B Technical Report
by: Aneja, Jyoti, et al.
Published: (2026)

Graph Unlearning with Efficient Partial Retraining
by: Zhang, Jiahao, et al.
Published: (2024)

On the Impossibility of Retrain Equivalence in Machine Unlearning
by: Yu, Jiatong, et al.
Published: (2025)

Rethinking Explanation Evaluation under the Retraining Scheme
by: Cai, Yi, et al.
Published: (2025)

Reconstruct the Pruned Model without Any Retraining
by: Wang, Pingjie, et al.
Published: (2024)

On the Stability of Iterative Retraining of Generative Models on their own Data
by: Bertrand, Quentin, et al.
Published: (2023)

The Unreasonable Effectiveness of Gaussian Score Approximation for Diffusion Models and its Applications
by: Wang, Binxu, et al.
Published: (2024)

A Multi-Criteria Automated MLOps Pipeline for Cost-Effective Cloud-Based Classifier Retraining in Response to Data Distribution Shifts
by: Katalay, Emmanuel K., et al.
Published: (2025)

Taming Hyperparameter Sensitivity in Data Attribution: Practical Selection Without Costly Retraining
by: Wang, Weiyi, et al.
Published: (2025)