🛰️ Daily AI Frontier
‹ back to 2026-09-08

How Does mHC Use Its Residual Streams? Selective Routing and Near-Identity Mixing

Research LLMs & Foundation Models

Ranking

Overall 74
Content 90
Popularity 37

Observed public metrics from 1 member.

Merged summary

TL;DR - An analysis of DeepSeek-V4-Flash finds that its four-stream mHC residual pathway uses selective routing and mostly near-identity mixing. The results suggest much of mHC’s theoretical flexibility is unnecessary at individual blocks, especially in later layers.

  • Attention and FFN sites effectively use about two streams on average, with the dominant stream changing across depth.
  • Stream representations remain directionally distinct, while cross-stream mixing is modest and concentrated in early layers.
  • Replacing late-layer mixers with identity raises C4 perplexity by only 1.9% without changing the six-task average; replacing early mixers raises perplexity by 41%.
  • Keeping only the three largest routing weights per token changes perplexity by at most 2.7% and the average task score by at most 0.4 points.

Sources (1)

How Does mHC Use Its Residual Streams? Selective Routing and Near-Identity Mixing

arXiv cs.LG Pengxiang Zhao, Xing Li, Xianzhi Yu, Wei Guo, Zhenhua Dong 2026-09-04 arXiv:2609.05309
Public signals Semantic Scholar citations 0 · Semantic Scholar influential citations 0
Providers: Hugging Face · N/A OpenAlex · N/A Publisher · N/A Semantic Scholar · Citations 0 · Influential citations 0 X · N/A Fetched 2026-09-10 14:16:31.594741 UTC

TL;DR - An analysis of DeepSeek-V4-Flash finds that its four-stream mHC residual pathway uses selective routing and mostly near-identity mixing. The results suggest much of mHC’s theoretical flexibility is unnecessary at individual blocks, especially in later layers.

  • Attention and FFN sites effectively use about two streams on average, with the dominant stream changing across depth.
  • Stream representations remain directionally distinct, while cross-stream mixing is modest and concentrated in early layers.
  • Replacing late-layer mixers with identity raises C4 perplexity by only 1.9% without changing the six-task average; replacing early mixers raises perplexity by 41%.
  • Keeping only the three largest routing weights per token changes perplexity by at most 2.7% and the average task score by at most 0.4 points.
item →