# HERO-MoE: Historical Expert Routing with Scale-Preserving Fusion

arXiv: https://arxiv.org/abs/2609.32581 · HTML: https://mlsanity.com/paper/2609.32581/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Expert routing strategies, one of 12 directions in Mixture-of-experts.
- Builds on 23 of its 45 references that are in the map.
- First posted on arXiv 26 Sep 2026.

## Placed in

- Mixture-of-experts / [Expert routing strategies](https://mlsanity.com/moe/expert-routing-strategies/)

## Builds on (most cited first, 23 of 23)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model](https://mlsanity.com/paper/2405.04434/) (2024)
- [Tulu 3: Pushing Frontiers in Open Language Model Post-Training](https://mlsanity.com/paper/2411.15124/) (2024)
- [DeepSeek-V3 Technical Report](https://mlsanity.com/paper/2412.19437/) (2024)
- [ST-MoE: Designing Stable and Transferable Sparse Expert Models](https://mlsanity.com/paper/2202.08906/) (2022)
- [Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts](https://mlsanity.com/paper/2408.15664/) (2024)
- [FastMoE: A Fast Mixture-of-Expert Training System](https://mlsanity.com/paper/2103.13262/) (2021)
- [MegaMath: Pushing the Limits of Open Math Corpora](https://mlsanity.com/paper/2504.02807/) (2025)
- [Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model](https://mlsanity.com/paper/2404.04167/) (2024)
- [Load Balancing Mixture of Experts with Similarity Preserving Routers](https://mlsanity.com/paper/2506.14038/) (2025)
- [Synergistic Intra- and Cross-Layer Regularization Losses for MoE Expert Specialization](https://mlsanity.com/paper/2602.14159/) (2026)
- [Routers Learn the Geometry of Their Experts: Geometric Coupling in Sparse Mixture-of-Experts](https://mlsanity.com/paper/2605.12476/) (2026)
- [AURA: A Diagnostic Framework for Tracking User Satisfaction of Interactive Planning Agents](https://mlsanity.com/paper/2505.01592/) (2025)
- [Evidence for Shared Routing Geometry and Dynamics in Sparse Mixture-of-Experts](https://mlsanity.com/paper/2609.02404/) (2026)
- [Multi-level context Modeling for consistent expert selection in Mixture-of-Experts](https://mlsanity.com/paper/2607.16427/) (2026)

## Abstract (arXiv)

Mixture-of-Experts (MoE) architectures have become a standard way to scale model capacity while keeping computation sparse, yet routing remains a key determinant of MoE quality and training behavior. Prior empirical studies suggest that MoE routing reflects input semantics and upstream computation across depth, but standard routers do not explicitly use the routing distributions produced by preceding layers. We propose HERO-MoE, Historical Expert ROuting with Scale-Preserving Fusion, a routing framework that injects historical routing priors into MoE routers by reusing detached, dense routing distributions collected from preceding MoE layers. The key idea is simple: HERO-MoE preserves the original token-conditioned routing branch and adds a residual historical routing contribution before the standard softmax and top-k dispatch. To stabilize this historical signal, HERO-MoE introduces a scale-preserving fusion mechanism that matches the magnitude of historical routing memory to the current hidden representation and accounts for the number of visible historical layers, without introducing an auxiliary routing loss or a fusion-specific tuning parameter. By reusing routing distributions already computed by preceding MoE layers, HERO-MoE improves training-loss reduction with modest end-to-end overhead. The resulting router remains compatible with standard sparse dispatch, including top-k and group-limited routing, and can be inserted into existing MoE backbones with minimal architectural changes. Experiments on an approximately 8B-parameter MoE model with 0.5B active parameters, trained from scratch on 100B tokens, show that HERO-MoE reduces the final loss from 1.6393 to 1.6184, while peak memory and FLOPs increase by only 0.44% and 0.64%, respectively.