iWSP: LLM Agents Build Interpretable Neural Nets by Programming Weights Directly
kalyan_kpl · x · 2026-11-20
A new paper introduces Interpretable Weight-Space Programming (iWSP), where LLM agents construct interpretable neural networks by directly programming weights and computations in code — no backpropagation or conventional training.
- Motivation: backprop-learned weights are hard to interpret; the paper tests whether agents can build accurate networks from explicitly programmed, human-readable computations.
- Loop: pick suitable computational operations for the task, write architecture/weights/features in code, evaluate with a fixed evaluator, and iteratively refine weights and computations based on results.
Related event: iWSP Lets LLM Agents Directly Program Neural Network Weights(2 posts)→
More from Research
- Image Models Now Ship Fine-tuned LLM Prompt Upsamplers — Why Not Fuse Them? — ostrisai · 2026-11-20
- PyTorch's ezyang on formalizing a trivial in-place kernel under PTX semantics — ezyang · 2026-11-20
- Scientist: Gemini 4 Neon designed innovative CAR receptors to kill cancer cells — DeryaTR_ · 2026-11-20
- New arXiv paper: Bregman power divergences turn KL queries into Euclidean NN in d+1 — FrnkNlsn · 2026-11-20
- Untrained Neural Networks Hit 86.4% AUROC on ADBench Anomaly Detection — ideal-lab-sejong · 2026-11-20
- NAVER's Unfinished-Bench: Persistent Agents Break Hidden Rules 43.6% of the Time — naver-ai · 2026-11-20