MAttr technical details: differentiable sigmoid top-k merges causal interventions and mask learning
aryaman2020 · x · 2026-09-24
Technical notes on MAttr: it combines the strengths of causal interventions, gradient-based attributions and mask learning, using a simple differentiable sigmoid top-k operator (no sparsity loss, no straight-through tricks) to parametrize causal interventions. Randomizing the top-k budget during training induces an attribution ranking; at eval time you simply set k to any desired sparsity.
More from Research
- Autonomous agent edits just 1,024 weights to lift Qwen3.5 0.8B from 7.4% to 37.4% on MATH-500 — VictorSanh · 2026-11-12
- Gait dataset labeled by encounter, spanning pre-op to 3 months post-TKA — zawthet · 2026-11-12
- Walking alignment measured 2.5-3.1° more neutral than X-ray, iPad gait study finds — zawthet · 2026-11-12
- Dynamic coronal alignment is a biomechanically distinct measure, may aid TKA planning — zawthet · 2026-11-12
- 28% of knees discordant between static and dynamic HKAA, valgus knees at 5.4x odds — zawthet · 2026-11-12
- iPad-based gait analysis presented at AAHKS: 110 knees, standing X-ray vs walking alignment — zawthet · 2026-11-12