Open-Source LLM Helps Safeguard Text Generation Prompts and Responses (corp.roblox.com) 2 points by moneil971 11mo ago ↗ HN
[–] moneil971 11mo ago ↗ Roblox’s LLM is currently outperforming popular LLM guardrail models on standard benchmarks. Roblox open-sourced both the LLM weights and the RoGuard-Eval benchmarking dataset.
1 comment
[ 3.1 ms ] story [ 12.9 ms ] thread