# Information-Theoretic Decoupled Prompt Tuning for Continual Learning

arXiv: https://arxiv.org/abs/2609.26257 · HTML: https://mlsanity.com/paper/2609.26257/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 50 references that are in the map.
- First posted on arXiv 13 Aug 2026.

## Builds on (most cited first, 10 of 10)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Experience Replay for Continual Learning](https://mlsanity.com/paper/1811.11682/) (2018)
- [Deep Variational Information Bottleneck](https://mlsanity.com/paper/1612.00410/) (2016)
- [On Tiny Episodic Memories in Continual Learning](https://mlsanity.com/paper/1902.10486/) (2019)
- [CLIP model is an Efficient Continual Learner](https://mlsanity.com/paper/2210.03114/) (2022)
- [A Unified Approach to Domain Incremental Learning with Memory: Theory and Algorithm](https://mlsanity.com/paper/2310.12244/) (2023)
- [External Knowledge Injection for CLIP-Based Class-Incremental Learning](https://mlsanity.com/paper/2503.08510/) (2025)
- [Adaptive Retention & Correction: Test-Time Training for Continual Learning](https://mlsanity.com/paper/2405.14318/) (2024)
- [Preserving Cross-Modal Consistency for CLIP-based Class-Incremental Learning](https://mlsanity.com/paper/2511.10974/) (2025)

## Abstract (arXiv)

Continual learning (CL) aims to incrementally acquire knowledge from sequential data while avoiding catastrophic forgetting. Recently, prompt tuning has attracted increasing attention as an efficient approach for adapting pre-trained models to CL tasks. However, existing prompt design paradigms commonly suffer from retrieval dependence and classifier bias, which make model adaptation sensitive to prompt selection and bias predictions toward newly arrived classes. To address these challenges, we propose Decoupled Prompt Tuning for Continual Learning (DPT4CL), which decouples the CLIP textual prompt into a task-shared prompt distribution and class-specific prompts. The task-shared prompt distribution is derived by optimizing an Information Bottleneck objective to facilitate cross-task knowledge transfer and alleviate classifier bias, while class-specific prompts enhance inter-class separability without relying on explicit prompt retrieval. Furthermore, we establish a unified excess risk bound from an information-theoretic perspective, providing theoretical support for the robust generalization and forgetting mitigation of the proposed framework. Extensive experiments on standard CL benchmarks demonstrate that DPT4CL achieves state-of-the-art performance. The source code is available at https://github.com/Cloudfly-Z/DPT4CL