Research map / Continual learning / Class-incremental
TailAdapt: Heavy-Tailed Sparse Variational Adaptation for Long-Tailed Class Incremental Learning
Abhishek Kumar Sinha, Nitant Dube, Soma Biswas
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes TailAdapt, a heavy-tailed sparse variational adaptation method for long-tailed class-incremental learning that addresses asymmetric likelihood signals between head and tail classes.
Problem
Long-tailed class-incremental learning (LT-CIL), where severe class imbalance lets head classes dominate gradient updates and tail classes stay near the prior mean because they contribute insufficient likelihood evidence within each session.
Approach
Identifies an asymmetry in per-session likelihood signals between head and tail classes and uses heavy-tailed sparse variational adaptation to counter prior regularization suppressing tail-class parameters. Details beyond the abstract are not available.
Closest papers by meaning
- Long-Tail Class Incremental Learning via Independent Sub-prototype Construction (CVPR 2024 · similarity 0.88)
- Variational Learning of Fractional Posteriors (ICML 2025 · similarity 0.87)
- LT-Soups: Bridging Head and Tail Classes via Subsampled Model Soups (NeurIPS 2025 · similarity 0.87)
- $t^3$-Variational Autoencoder: Learning Heavy-tailed Data with Student's t and Power Divergence (ICLR 2024 · similarity 0.87)
- SD-LoRA: Scalable Decoupled Low-Rank Adaptation for Class Incremental Learning (ICLR 2025 · similarity 0.86)
- Long-Tailed Learning as Multi-Objective Optimization (AAAI 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
