Steven Hoi, Binh Nguyen, Chenghao Liu, Trungtin Nguyen, Quang Pham, Xiaoli Li, Savitha Ramasamy, Thanh-Nam Doan, Khiem Le, Giang Do, Chi City
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
By routing input tokens to only a few split experts, Sparse Mixture-of-Experts has enabled efficient training of large language models. Recent findings suggest that fixing the routers can achieve competitive performance by alleviating the collapsing problem, where all experts eventually learn similar representations. However, this strategy has two key limitations: (i) the policy derived from random routers might be suboptimal, and (ii) it requires extensive resources during training and evaluation, leading to limited efficiency gains. This work introduces HyperRouter, which dynamically generates the router's parameters through a fixed hypernetwork and trainable embeddings to achieve a balance between training the routers and freezing them to learn an improved routing policy. Extensive experiments across a wide range of tasks demonstrate the superior performance and efficiency gains of HyperRouter compared to existing routing methods. Our implementation is publicly available at https://github.com/ giangdip2410/HyperRouter.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2312.07035")
get_code_for_paper("2312.07035")
have("2312.07035")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2312.07035.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2312.07035)
Connect an agent — have() is free.