Skip to content

[Off-topic] Question for Damai Dai / Chengqi Deng: has profile-based MoE routing been explored? #77

Description

@washingtoneimae-dot

I noticed this repo has recent activity from Damai Dai and Chengqi Deng — the same researchers behind DeepSeekMoE's expert routing architecture.

Quick question on MoE routing: has the team considered replacing the learned router W_r·x with profile-based routing? Each expert carries a calibrated benchmark vector. Router uses cosine similarity instead of learned weights. Zero router parameters (cosine sim is pure math). Experts become swappable post-training (update one array, no retraining).

At V3 scale this reduces routing params from ~106M to ~500K. At V4 scale (~384 experts) it's ~192M vs ~500K. The bias balancing mechanism remains compatible — just add b_i to cosine similarity scores before top-K.

We have a working implementation proving this (99.9% routing accuracy, 38× swap isolation, identical inference speed): https://github.com/washingtoneimae-dot/profile-moe

Would love a technical opinion from the people who built the original MoE architecture.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions