← back to paper

Review history

arxiv: 2608.09740 · 2 revisions

Security Tests as Executable Specifications for LLM Code Generation: Benefits, Trade-offs, and Coverage Limits

1. 2026-08-14 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 6.0
   124504 ms 14031 in 12816 out 2026-08-14T04:14:05.181781+00:00
2. 2026-08-11 ACCEPT HIGH v1.4.0-daily-deepseek-medium novelty 6.0
   81721 ms 14020 in 9935 out 2026-08-11T11:39:15.849971+00:00