Domingos on iWSP paper: just use the program instead of writing it into LLM weights
pmddomingos · x · 2026-11-18
Pedro Domingos pushes back on the iWSP paper (Microsoft Research et al.), which has an LLM agent hand-craft interpretable models by writing weights in code without backpropagation. His take: skip the weights entirely and just use the program — more interpretable and less wasted effort. iWSP reports outpredicting interpretable baselines on visual-cortex encoding tasks.
More from Models
- Astra 6.2 plays detective: finds URL on a projector screen in hours of B-roll — gabrielchua · 2026-11-18
- Qwen 3.8 27B vs Flash-Next vs DeepSeek V4.1 Flash: one-shot landing page build-off — jesdga95 · 2026-11-18
- GPT-6.2 Astra plays RimWorld for 250 in-game days and builds a decent colony — jxnlco · 2026-11-18
- Polymarket claims GPT-6 Cyber uncovered 35 security flaws across Apple iPhones and Macs — Polymarket · 2026-11-18
- OpenAI ships GPT-6.2 Astra: 22-point ExploitBench jump, tighter cyber safeguards — charliermarsh · 2026-11-18
- Halocline inference engine hits 148-182 tok/s on Qwen-Flash-Next with KLD 0.082 — Distinct-Pie2389 · 2026-11-18