Yoav Goldberg Weighs In on Weight-Interpretation Experiments: Fixing Models by Editing Biases Is More Stunning Than Writ…
yoavgo · x · 2026-11-03
AI researcher Yoav Goldberg (yoavgo) commented on a recent wave of experiments that directly manipulate network weights. What he finds most compelling is research on "understanding what a network is doing purely from its weights." By contrast, "hand-writing weights from scratch to accomplish a task" doesn't impress him as much—his students once hand-wrote weights for small networks on an exam, and this is essentially the same thing done by computers at a much larger scale.
More from Research
- After two silent grant rejections, researcher posts federated learning paper on arXiv — wandedob · 2026-11-03
- Scale AI's READY framework: identical-accuracy agents need 39.2% vs 29.6% human review — demian_ai · 2026-11-03
- LLMs can natively edit neural network weights: 0.016% of changes boost small model math 6x — Dr_Singularity · 2026-11-03
- Developer warns: do NOT use weight-space programming — djcows · 2026-11-03
- Lean 4 proof shows every finite simple group is generated by two elements — geoffreyirving · 2026-11-03
- Musk: diffusion wins on any bitstream; CoBit applies one diffusion recipe to text, speech, images, proteins — LucaAmb · 2026-11-03