Research map / Continual learning / Continual fine-tuning of LLMs
DeLo: Dual Decomposed Low-Rank Experts Collaboration for Continual Missing Modality Learning
Xiwei Liu, Yulong Li, Feilong Tang, Imran Razzak
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add Cross-Modal Guided Routing (ablation) | DeLo · UPMC-Food101-CMML (eta=70%) | AP: 48.84 -> 67.42 | Improved |
| add Modality-Specific Factor Pool (separate vision and text pools) (ablation) | DeLo · UPMC-Food101-CMML (eta=70%) | AP: 56.52 -> 67.42 | Improved |
| use dynamic decomposed LoRA experts instead of static LoRA (ablation) | DeLo · UPMC-Food101-CMML (eta=70%) | AP: 60.39 -> 67.42 | Improved |
| increase selected rank from 2 to 4 (ablation) | DeLo (E=16) · UPMC-Food101-CMML (eta=70%) | AP: 63.78 -> 67.42 | Improved |
| increase selected rank from 4 to 8 (ablation) | DeLo (E=16) · UPMC-Food101-CMML (eta=70%) | AP: 67.42 -> 67.11 | No clear effect |
| increase factor pool from E=16 to E=32 at r=4 (ablation) | DeLo · UPMC-Food101-CMML (eta=70%) | AP: 67.42 -> 68.03 | Improved |
| add consistency loss to classification loss (ablation) | DeLo · UPMC-Food101-CMML (eta=70%) | AP: 61.95 -> 65.15 | Improved |
| add alignment loss to classification loss (ablation) | DeLo · UPMC-Food101-CMML (eta=70%) | AP: 61.95 -> 65.88 | Improved |
| add both alignment and consistency losses (ablation) | DeLo · UPMC-Food101-CMML (eta=70%) | AP: 61.95 -> 67.42 | Improved |
| use DeLo instead of RebQ | RebQ · UPMC-Food101-CMML (eta=90%, image 100%, text 10%) | AP: 48.15 -> 67.30 | Improved |
| use DeLo instead of RebQ | RebQ · MM-IMDb-CMML (eta=90%, image 100%, text 10%) | AP: 21.37 -> 26.11 | Improved |
| use DeLo instead of RebQ (forgetting) | RebQ · MM-IMDb-CMML (eta=90%, image 100%, text 10%) | FG: 23.66 -> 35.11 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DualPrompt — DualPrompt: Complementary Prompting for Rehearsal-Free Continual Learning (ECCV 2022) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Mitigating Intra- and Inter-modal Forgetting in Continual Learning of Unified Multimodal Models (NeurIPS 2025 · similarity 0.90)
- Do More Modalities Always Help? A Geometric Perspective on Missing-Modality Robustness (NeurIPS 2026 · similarity 0.89)
- FLAME: Adaptive Mixture-of-Experts for Continual Multimodal Multi-Task Learning (NeurIPS 2026 · similarity 0.89)
- Retrieval-Augmented Dynamic Prompt Tuning for Incomplete Multimodal Learning (AAAI 2025 · similarity 0.88)
- Knowledge Bridger: Towards Training-Free Missing Modality Completion (CVPR 2025 · similarity 0.88)
- Dynamic Mixture of Curriculum LoRA Experts for Continual Multimodal Instruction Tuning (ICML 2025 · similarity 0.87)
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.
