BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Allsikt//Article Deadline//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:bc9f954898f67016b26d25adfd2aaf005b92aef4@allsikt.tech
DTSTAMP:20260722T142247Z
DTSTART;VALUE=DATE:20260722
DTEND;VALUE=DATE:20260723
SUMMARY:ProgramBench: Can Language Models Rebuild Programs From Scratch?
DESCRIPTION:Explore ProgramBench to benchmark your LMs on end‑to‑end program synthesis\; evaluate against 200 tasks covering CLI tools\, FFmpeg\, SQLite\, PHP\, and note that current models only achieve 95 % pass on 3 % of tasks.\n\nSource: Hacker News (front page)\nOpen: https://allsikt.se/article/programbench-can-language-models-rebuild-programs-from-scratch-a1ee4ece
URL:https://allsikt.se/article/programbench-can-language-models-rebuild-programs-from-scratch-a1ee4ece
STATUS:CONFIRMED
TRANSP:TRANSPARENT
END:VEVENT
END:VCALENDAR