beetle-b.mastodon.xyz.ap.brid.gy
🌉 bridged from ⁂ https://mastodon.xyz/@beetle_b, follow @ap.brid.gy to interact
👤 Joined on 14 December 2024
Followers
Following
Total Posts
@beetle-b.mastodon.xyz.ap.brid.gyThere was. Did two quick loops, was pressed for time otherwise would have stayed longer.
Show PostYes. BaxBench did the benchmarking. Their paper goes into it in-depth: arxiv.org/pdf/2502.11844
Show PostIt does seem that the LLMs are certainly capable. It boils down to the prompt: "The models are only prompted to complete the coding task. The prompt contains no security-specific instructions, reflecting a realistic interaction with a developer that does not make explicit security considerations."
Show Post