arXiv Machine Learning September 25, 2026 By Di Fang, Yinan Zhu, Zhiping Lin, Cen Chen, Ziqian Zeng, Huiping Zhuang

AIR: Analytic Imbalance Rectifier for Continual Learning

Read the original on arXiv Machine Learning →

The paper introduces AIR, an analytic imbalance rectifier designed for continual learning scenarios where data streams are evolving and imbalanced. AIR operates without exemplars, using a frozen backbone for feature extraction and a closed‑form incremental classifier that incorporates a class‑weighted ridge objective. It employs an analytic reweighting module to equalize sample weights across classes, achieving significant improvements in accuracy and macro F1 over 28 baseline methods in long‑tailed class‑incremental learning and over 15 baselines in the Si‑Blurry setting with recurring classes.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

agents

arXiv Machine Learning

Aug 18

Free-Flow Class-Incremental Learning: Towards Robust CIL under Variable Class Arrivals

arXiv:2604. 02765v2 Announce Type: replace Abstract: Class-incremental learning (CIL) is commonly evaluated under predefined schedules with fixed or nearly equal class increments, leaving irregular class-arrival scenarios underexplored.

By Zhiming Xu, Baile Xu, Jian Zhao, Furao Shen, Suorong Yang

safety

arXiv Machine Learning

Jul 10

Are Current Continual Learning Methods Truly Agnostic? Introducing OPRE, a Step Toward Agnostic Continual Learning

arXiv:2511. 08226v2 Announce Type: replace Abstract: In order to achieve Continual Learning (CL), the problem of catastrophic forgetting, one that has plagued neural networks since their inception, must be overcome.

By Rapha\"el Bayle, Martial Mermillod, Robert M. French

benchmarks

arXiv AI

Jul 29

CIFNet: An Analytic Neural Learning Framework for Efficient and Calibrated Class-Incremental Learning

arXiv:2509. 11285v2 Announce Type: replace-cross Abstract: Class-Incremental Learning (CIL) in deep neural networks is conventionally framed as an iterative gradient-based optimization problem, incurring high computational cost, hyperparameter sensitivity, and risk of catastrophic forgetting.

By Alejandro Dopico-Castro, Oscar Fontenla-Romero, Bertha Guijarro-Berdi\~nas, Amparo Alonso-Betanzos

rag diffusion safety

arXiv Machine Learning

Aug 3

TOOD: Task-Aware Out-of-Distribution Score Calibration for Continual Learners

arXiv:2607. 29592v1 Announce Type: cross Abstract: The primary challenge of continual learning (CL) systems is to learn new tasks while remaining performant on previously learned tasks.

By Mostafa ElAraby, Samer B. Nashed, Liam Paull

arXiv AI

Jul 22

Parameter-Efficient Continual Fine-Tuning: A Survey

arXiv:2504. 13822v3 Announce Type: replace-cross Abstract: The emergence of large pre-trained networks has revolutionized the AI field, unlocking new possibilities and achieving unprecedented performance.

By Eric Nuertey Coleman, Luigi Quarantiello, Ziyue Liu, Qinwen Yang, Samrat Mukherjee, Julio Hurtado, Vincenzo Lomonaco

fine-tuning benchmarks

arXiv Machine Learning

Jun 11

RCAP: Robust, Class-Aware, Probabilistic Dynamic Dataset Pruning

arXiv:2606. 11761v1 Announce Type: new Abstract: Dynamic data pruning techniques aim to reduce computational cost while minimizing information loss by periodically selecting representative subsets of input data during model training.

By Atif Hassan, Swanand Khare, Jiaul H. Paik

fine-tuning efficiency benchmarks

One email a morning, machine-written

Related stories

Free-Flow Class-Incremental Learning: Towards Robust CIL under Variable Class Arrivals

Are Current Continual Learning Methods Truly Agnostic? Introducing OPRE, a Step Toward Agnostic Continual Learning

CIFNet: An Analytic Neural Learning Framework for Efficient and Calibrated Class-Incremental Learning

TOOD: Task-Aware Out-of-Distribution Score Calibration for Continual Learners

Parameter-Efficient Continual Fine-Tuning: A Survey

RCAP: Robust, Class-Aware, Probabilistic Dynamic Dataset Pruning