Pith. sign in

REVIEW 6 cited by

Don't Stop Learning: Towards Continual Learning for the CLIP Model

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2207.09248 v2 pith:2P6ZXHSE submitted 2022-07-19 cs.CV

Don't Stop Learning: Towards Continual Learning for the CLIP Model

classification cs.CV
keywords cliplearningmodelcontinualforgettingissueimage-textcatastrophic
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

The Contrastive Language-Image Pre-training (CLIP) Model is a recently proposed large-scale pre-train model which attracts increasing attention in the computer vision community. Benefiting from its gigantic image-text training set, the CLIP model has learned outstanding capabilities in zero-shot learning and image-text matching. To boost the recognition performance of CLIP on some target visual concepts, it is often desirable to further update the CLIP model by fine-tuning some classes-of-interest on extra training data. This operation, however, raises an important concern: will the update hurt the zero-shot learning or image-text matching capability of the CLIP, i.e., the catastrophic forgetting issue? If yes, could existing continual learning algorithms be adapted to alleviate the risk of catastrophic forgetting? To answer these questions, this work conducts a systemic study on the continual learning issue of the CLIP model. We construct evaluation protocols to measure the impact of fine-tuning updates and explore different ways to upgrade existing continual learning methods to mitigate the forgetting issue of the CLIP model. Our study reveals the particular challenges of CLIP continual learning problem and lays a foundation for further researches. Moreover, we propose a new algorithm, dubbed Learning without Forgetting via Replayed Vocabulary (VR-LwF), which shows exact effectiveness for alleviating the forgetting issue of the CLIP model.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. StructAlign: Structured Cross-Modal Alignment for Continual Text-to-Video Retrieval

    cs.CV 2026-01 unverdicted novelty 7.0

    StructAlign uses simplex ETF geometry and cross-modal relation-preserving losses to mitigate intra- and cross-modal feature drift in continual text-to-video retrieval.

  2. Continual Learning for VLMs: A Survey and Taxonomy Beyond Forgetting

    cs.CV 2025-08 unverdicted novelty 7.0

    The paper offers a comprehensive survey and proposes a new taxonomy for continual learning strategies in VLMs and MLLMs to combat catastrophic forgetting beyond traditional methods.

  3. CMAP: Cross-Modal Adaptive Prompting for Multi-Domain Task-Incremental Learning

    cs.CV 2026-05 unverdicted novelty 6.0

    CMAP uses text-space task routing with frozen CLIP prototypes, multi-prototype cross-modal confidence, and symmetric gating to achieve new state-of-the-art results on the MTIL benchmark with 2.5M parameters.

  4. DIMoE-Adapters: Dynamic Expert Evolution for Continual Learning in Vision-Language Models

    cs.CV 2026-05 unverdicted novelty 6.0

    DIMoE-Adapters uses self-calibrated expert evolution and prototype-guided selection to dynamically grow and allocate experts, outperforming prior continual learning methods on vision-language models.

  5. Little by Little: Continual Learning via Incremental Mixture of Rank-1 Associative Memory Experts

    cs.LG 2025-06 unverdicted novelty 6.0

    MoRAM frames continual learning as incremental addition of rank-1 adapters viewed as self-activating key-value associative memory units in a mixture-of-experts setup.

  6. iGSP:Implicit Gradient Subspace Projection for Efficient Continual Learning of Vision-Language Models

    cs.CV 2026-05 unverdicted novelty 5.0

    iGSP uses implicit gradient subspace projection in two phases to enable efficient continual adaptation of vision-language models, claiming SOTA accuracy with 42.7% fewer trainable parameters and 86.9% less total param...