UCSB and Microsoft's ScholarEvolve generates code from academic papers to improve an agent's harness without retraining model weights. It raised Qwen3.5-27B's task completion rate to 81.4%. Here is ...
A preliminary AES experiment found that bioinformatics-inspired alignment recovered keys with up to 44% fewer power traces ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results