Harnessing Domain Specialists in Multimodal Mixture-of-Experts for Efficient Adaptation

arXiv:2610.02123 · cs.CV · Submitted 2026-10-01 · Read on arXiv

cs.CV

Submitted: 2026-10-01

Updated: 2026-10-01

Code: https://github.com/damianomarsili/ExpertLens

Terminology

• Adaptation — 48× in this paper · explained in 1 episode
• Accuracy — 13× in this paper · explained in 1 episode
• Fine-Tuning — 13× in this paper · explained in 5 episodes
• Sparsity — 10× in this paper · explained in 4 episodes
• Benchmark — 9× in this paper · explained in 2 episodes
• Transformer — 8× in this paper · explained in 2 episodes
• Modularity — 7× in this paper · explained in 9 episodes
• Clustering — 5× in this paper · explained in 2 episodes
• Mixture-of-Experts (MoE) — 5× in this paper · explained in 7 episodes
• Alignment — 3× in this paper · explained in 1 episode
• Interpretability — 3× in this paper · explained in 15 episodes
• Ablation — 2× in this paper · explained in 1 episode
• Decode — 2× in this paper · explained in 2 episodes
• Foundation Model — 2× in this paper · explained in 3 episodes
• Foundation Models — 2× in this paper · explained in 7 episodes
• Mixture of Experts — 2× in this paper · explained in 2 episodes
• Mixture-of-Experts (MoE) Models — 2× in this paper · explained in 3 episodes
• Multimodal Reasoning — 2× in this paper · explained in 5 episodes
• Robustness — 2× in this paper · explained in 5 episodes
• Adam — 1× in this paper · explained in 2 episodes
• Aggregation — 1× in this paper · explained in 1 episode
• Conference on Computational Natural Language Learning — 1× in this paper
• Interpretability of Mixture-of-Experts — 1× in this paper
• Parameter-Efficient Fine-Tuning — 1× in this paper · explained in 2 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Reward Function — 1× in this paper · explained in 2 episodes
• Semantic Similarity — 1× in this paper · explained in 2 episodes
• Steering — 1× in this paper · explained in 2 episodes
• Verifier — 1× in this paper · explained in 2 episodes

Sources

• Qwen3-VL Technical Report
• Eliciting Latent Predictions from Transformers with the Tuned Lens
• MoE Lens -- An Expert Is All You Need
• Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
• Training Verifiers to Solve Math Word Problems
• Towards Automated Circuit Discovery for Mechanistic Interpretability
• DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models
• Steering MoE LLMs via Expert (De)Activation
• Gemma 4 Technical Report
• Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space
• PathVQA: 30000+ Questions for Medical Visual Question Answering
• What Gets Activated: Uncovering Domain and Driver Experts in MoE Language Models
• Mixtral of Experts
• What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams
• PubMedQA: A Dataset for Biomedical Research Question Answering
• GeomVerse: A Systematic Evaluation of Large Models for Geometric Reasoning
• Kimi K3: Open Frontier Intelligence
• Kimi-VL Technical Report
• GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding
• Decoding Knowledge Attribution in Mixture-of-Experts: A Framework of Basic-Refinement Collaboration and Efficiency Analysis

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models