Neural Networks Construct Functions in Context
andrew_n_carr · x · 2026-07-13
The author argues that the traditional view of neural networks as "learning a function" is no longer accurate.
A more apt description:
- Important neural networks don't directly learn a static function;
- They more like construct functions based on context;
- The actual mechanism is closer to a "function factory", and this process can be done by in-context circuits without gradients.
This is a conceptual reframing emphasizing the central role of context in inference.
More from Research
- Structural ensembles beat single predictions in TCR:pMHC generalization study — quaidmorris · 2026-07-22
- RSS launches under OMSF to push structural biology data modeling at scale — MoAlQuraishi · 2026-07-22
- enFoldX tops 8 neoantigen scans and an unseen-peptide benchmark — quaidmorris · 2026-07-22
- enFoldX reaches AUC 0.82 on human VDJdb and transfers to mouse at 0.76 — quaidmorris · 2026-07-22
- enFoldX gains accuracy as AF3 ensemble disagreement rises for non-binders — quaidmorris · 2026-07-22
- A 3D ray plot shows how hard this Jacobian counterexample is to read — moultano · 2026-07-22