Mixture-of-experts vision-language models (MoE-VLMs) increase model capacity with sparse expert activation, yet deployment requires storing the full expert pool. Training-free expert merging reduces this burden, and many routing-based methods aggregate routing statistics across all tokens to determine merge compatibility.
arXiv:2608. 04454v1 Announce Type: cross Abstract: Mixture-of-experts vision-language models (MoE-VLMs) increase model capacity with sparse expert activation, yet deployment requires storing the full expert pool.
arXiv:2511. 08972v2 Announce Type: replace Abstract: Sparse Mixture-of-Experts (SMoE) models are scalable and computationally efficient, enabling large increases in model capacity with limited inference overhead.
arXiv:2608.24763v1 Announce Type: cross Abstract: Procedural video-language models must solve heterogeneous tasks from the same visual evidence, including action recognition, forecasting, and procedu...
The paper introduces L2R, a routing framework for Mixture-of-Experts models that reshapes the routing space into a shared low‑rank latent space and employs Saturated Inner‑Product Scoring to control Lipschitz behavior, resulting in smoother and more stable routing geometry. It also adds a parameter‑efficient multi‑anchor routing mechanism to increase expert expressiveness. Experiments on an OLMoE‑based language model and a ViT‑based ImageNet setting demonstrate improved overall performance and better routing geometry and expert discrimination.
arXiv:2608. 06396v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) language models route each token through a small subset of experts, making routing patterns useful for identifying task-relevant experts during downstream adaptation.
arXiv:2609.38823v1 Announce Type: new Abstract: Multimodal mixture-of-experts (MoE) models combine sparse expert activation with visual-language capabilities, yet their inference remains costly becau...
arXiv:2608. 10392v1 Announce Type: new Abstract: Mixture-of-experts (MoE) models have recently moved beyond routing a fixed number of complete experts.
arXiv:2604. 00421v2 Announce Type: replace Abstract: Mixture-of-Experts (MoE) layers increase model capacity by activating only a small subset of experts per token, and typically rely on a learned router to map hidden states to expert assignments.
IntBMoE introduces a block‑conditioned mixture‑of‑experts that decouples participation, execution, and materialization by combining dense expert composition with sparse block execution. Each internal layer uses a lightweight hypernetwork to merge all expert bases into a single composed expert, while a router selects only a few blocks per token, keeping compute and memory costs low. Experiments on image classification, language modeling, and sequential recommendation demonstrate consistent performance gains, and the model is deployed in AMap’s generative recommendation system, improving UVCTR by 2.4% in online A/B tests.
The paper introduces Role-Conditioned Sub-Token Routing (RoleSub), a method that compresses the value representations of retained tokens in Vision‑Language‑Action models. By partitioning each token’s value into orthogonal groups and routing them based on token features, latent roles, and language context, RoleSub can compress both visual and language representations without discarding tokens. Experiments on OpenVLA‑OFT‑7B show that, at matched visual‑KV budgets, RoleSub outperforms token‑only control in most settings and can reduce total KV to 9.2–11.3% of the original while maintaining strong control performance.
The paper introduces Role-Conditioned Sub-Token Routing (RoleSub), a method that compresses the value representations of retained tokens in Vision‑Language‑Action models. By partitioning each token’s value into orthogonal groups and routing them based on token features, a latent role, and language context, RoleSub can also compress language values. Experiments on OpenVLA‑OFT‑7B show that, at matched visual‑KV budgets, RoleSub outperforms token‑only control in most settings and reduces total KV to 9.2–11.3% of the original while maintaining strong control performance.
The paper investigates whether the sparsity of Mixture-of-Experts (MoE) models leads to intrinsic semantic organization across modalities and domains. It shows that experts naturally specialize semantically even without explicit modular training. The authors propose ExpertLens, a data‑free method that decodes router weights to identify domain‑specialized experts, enabling selective fine‑tuning that matches or exceeds full fine‑tuning while updating only 21.7–47.0% of parameters and achieving a 4.0× speedup, outperforming LoRA in both performance and efficiency.