fordycelab proposes benchmarking binder-design algorithms with Kd measurements on 24,000-VHH libraries
anshulkundaje · x · 2026-11-08
An experimentalist proposes benchmarking current binder-design algorithms by quantifying Kds for libraries of 24,000 VHHs against several targets, asking the community which targets and algorithms to nominate and what results to expect — a rare wet-lab-anchored evaluation discussion in AI protein design.
More from Research
- MuJoCo 4.0 ships with speculative contacts, permanent soft-body deformation and MuJoCo Studio — ZoubinGhahrama1 · 2026-11-08
- mitsuhiko: djs' JS "heap dumps" may beat real dumps for durable coding agents — mitsuhiko · 2026-11-08
- Salesforce paper: RL-trained experts merged via OPD save 42% FLOPs vs external-teacher distillation — rohanpaul_ai · 2026-11-08
- Albert Gu: Bitter Lesson isn't 'just scale Transformers' — algorithmic innovation matters — sewon__min · 2026-11-08
- Autoware open-sources an off-road driving simulator built on NVIDIA Isaac Sim — rsasaki0109 · 2026-11-08
- The meta-question as agent swarms scale: can models beat gradient descent at its own game? — JensHonack · 2026-11-08