🛰️ Daily AI Frontier
‹ back to 2026-09-21

Abstention and Noise Filtering: Two Missing Primitives of Softmax Attention

Research LLMs & Foundation Models

Ranking

Overall 82
Content 95
Popularity N/A

No observed public metrics; popularity remains neutral/archived.

Merged summary

TL;DR - This paper argues that value-pathway gates improve softmax attention by adding two missing capabilities: abstaining from producing an output and filtering noisy residual-stream features. Experiments across 10M–350M parameter models show that combining both mechanisms consistently performs best with negligible parameter overhead.

  • A learned per-head sink logit enables abstention, while per-value gating suppresses interference from superposed features.
  • Abstention provides most of the gating benefit at 10M parameters, but its impact declines as model size increases.
  • Noise filtering becomes more beneficial with scale and accounts for most of the improvement at 350M parameters.
  • Controlled interference experiments confirm the filtering effect and reveal distinct blind spots in each gate type; both remain compatible with key-value caching.

Sources (1)

Abstention and Noise Filtering: Two Missing Primitives of Softmax Attention

arXiv cs.LG Richard Zhe Wang 2026-09-18 arXiv:2609.22005
Public signals N/A
Providers: Hugging Face · N/A OpenAlex · N/A Publisher · N/A Semantic Scholar · N/A X · N/A Fetched 2026-09-26 14:14:55.213947 UTC

TL;DR - This paper argues that value-pathway gates improve softmax attention by adding two missing capabilities: abstaining from producing an output and filtering noisy residual-stream features. Experiments across 10M–350M parameter models show that combining both mechanisms consistently performs best with negligible parameter overhead.

  • A learned per-head sink logit enables abstention, while per-value gating suppresses interference from superposed features.
  • Abstention provides most of the gating benefit at 10M parameters, but its impact declines as model size increases.
  • Noise filtering becomes more beneficial with scale and accounts for most of the improvement at 350M parameters.
  • Controlled interference experiments confirm the filtering effect and reveal distinct blind spots in each gate type; both remain compatible with key-value caching.
item →