AI Paper Atlas

Research map / Continual learning / Continual fine-tuning of LLMs

D-VLAM: Differential Vision and Language Mixing for Rehearsal Free Continual Learning

Muhammad Anwar Ma'sum, Mohsen Guizani, Waseem Ullah

ECCV 2026 · Continual learning › Continual fine-tuning of LLMs · approach: Mixture-of-experts for multimodal continual tuning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add differential prefix tuning (lambda_t + DPT) (ablation)D-VLAM · ImageNet-RFAA: 85.5 -> 86.5Improved
add language-modality prompt generator Gt(E) (ablation)D-VLAM · ImageNet-RFAA: 82.2 -> 85.5Improved
add class-level keys and loss (ablation)D-VLAM · ImageNet-RFAA: 76.8 -> 82.2Improved
add task-level key and task-wise generators (ablation)D-VLAM · ImageNet-RFAA: 71.9 -> 76.8Improved
add vision prompt generator G(x) (ablation)fine-tuning · ImageNet-RFAA: 59.5 -> 71.9Improved
add inter-task loss (ablation)fine-tuning · ImageNet-RFFM: FAA 58.5 -> 59.5; FFM 5.9 -> 7.1Mixed
complementary-LV or aggregate instead of complementary-VL mixing (ablation)D-VLAM · ImageNet-RFAA: 86.5 vs 86.1 / 86.4No clear effect
add DPT to each mixing configuration (ablation)D-VLAM · ImageNet-RFAA: 85.4 -> 86.1 (Comp-LV)Improved
use D-VLAM instead of LEAPGenlanguage-guided prompt CL · ImageNet-R 10-taskFAA: 83.18 -> 86.54Improved
D-VLAM vs LEAPGen training timeD-VLAM · ImageNet-R 10-tasktraining time (h): 0.655 -> 0.725Hurt
use class-name text and lightweight generators instead of larger generatorsD-VLAM · ImageNet-R 10-tasktrainable params: 6.35 M -> 0.16 M (vs LEAPGen)Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy