Laying out DeepLIFT as a standalone computation graph: from sequence to attributions directly
jmschreiber91 · x · 2026-10-26
Continuing the thread, jmschreiber91 describes the fused implementation: instead of general-purpose hooks, DeepLIFT is laid out directly on a specific Cherimoya model as an entirely separate computation graph — input changes from sequence→prediction to sequence→attributions, letting PyTorch handle op fusion and optimization.
Related event: Developer rewrites DeepLIFT as a standalone computation graph(2 posts)→
More from Research
- 'Vibe verification': ~60% of AI-written formal specs prove the wrong property — basedjensen · 2026-10-26
- Ito: a browser that detects browsing intent and auto-organizes tabs into Flows — andrewkuznet · 2026-10-26
- Cornell study says brains process language as linear sequences, challenging Chomsky's hierarchy — anirbanbandyo · 2026-10-26
- Hi3D V1.0 open-sourced with 2M training assets and 90% slicing success rate — kimmonismus · 2026-10-26
- Ortet evaluates 34 open-weight models on 50 health tasks, finds general leaderboards insufficient — elmanmansimov · 2026-10-26
- HackBench injects reward hacks into agent trajectories to stress-test monitors — DanielKhashabi · 2026-10-26