DeepLIFT pitched as a faster, cleaner alternative to noisy ISM attribution
anshulkundaje · x · 2026-11-18
A reshared post promotes DeepLIFT as an attribution method, complaining that ISM is slow and grad×input is noisy. It links to a talk by @valehamiri2, @rrastogi02 and Salil Deshpande at MLCB, plus a podcast introducing interpretation methods in detail — a practical reference for researchers doing model attribution.
More from Research
- 27B Model Rewrites Its Own Weights Without Training or Gradients, Paper Claims — udmrzn · 2026-11-18
- From 'impossible' to a working Bluetooth beacon: open-source EDA chip design is real — IgorCarron · 2026-11-18
- AURA 2.0: a streaming video LLM that speaks first, calls tools, and grounds visually — Shuhan Li · 2026-11-18
- Orabona: the classic preconditioning story for full-matrix AdaGrad doesn't hold up — bremen79 · 2026-11-18
- StellarEdge Auriga report: verifiable execution harness hits 57.74% on 42 RoboDojo tasks — jiqizhixin · 2026-11-18
- AHA! Weight-space programming without an external LLM lifts Qwen3.5-0.8B GSM8K from 14.0 to 71.2 — hbouammar · 2026-11-18