AI Paper Atlas

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

How to Teach Large Multimodal Models New Skills

Zhen Zhu, Yiming Gong, Yao Xiao, Yaoyao Liu, Derek Hoiem

ECCV 2026 · Continual learning › Continual fine-tuning of LLMs · approach: Mixture-of-experts for multimodal continual tuning · 8 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
tune only self-attention projections instead of the full LLM (ablation)LLaVA-OneVision-7B · Held-out suite (5-task sequence)held-out forgetting: -23.3 -> -0.6Improved
tune only MLP Gate&Up (freeze Down) instead of full MLP (ablation)LLaVA-OneVision-7B · Held-out suite (5-task sequence)held-out forgetting: -15.7 -> -2.1Improved
freeze Down projection in MLP tuning (ablation)LLaVA-OneVision-7B · Target tasks (5-task sequence)target overall: 15.5 -> 27.1Improved
also tune the vision encoder (vs LLM only) (ablation)LLaVA-OneVision-7B · Held-out suite (5-task sequence)held-out forgetting: -10.8Hurt
freeze W_O in self-attention tuning (SA Proj. QKV) (ablation)LLaVA-OneVision-7B · Target tasks (5-task sequence)target overall: 23.1 -> 14.5Hurt
tune full model instead of LLM only (ablation)LLaVA-OneVision-7B · Target tasks (5-task sequence)target learning: 31.8 -> 29.9No clear effect
tune self-attention projections instead of full LLM (ablation)LLaVA-NeXT (LLaMA-3 8B) · Held-out suite (5-task sequence)held-out forgetting: -35.9 -> -7.7Improved
tune self-attention projections instead of full LLM (ablation)Qwen2.5-VL-7B · Held-out suite (5-task sequence)held-out forgetting: -24.6 -> 0.6Improved
tune MLP Gate&Up instead of full MLP (ablation)Qwen2.5-VL-7B · Target tasks (5-task sequence)target overall: 13.9 -> 17.1Improved
tune MLP Gate&Up instead of full MLP (ablation)LLaVA-NeXT (LLaMA-3 8B) · Held-out suite (5-task sequence)held-out forgetting: -16.3 -> -8.7Improved
tune self-attention projections instead of full model (single task) (ablation)LLaVA-OneVision-7B · Held-out average (single-task fine-tuning)held-out change: -3.2 -> +0.0Improved
tune MLP instead of full LLM (single task) (ablation)LLaVA-OneVision-7B · Held-out average (single-task fine-tuning)held-out change: -2.8 -> -1.4Improved

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