[Linkpost] “Guardian Angels: LLM Personalization for Productivity and Security” by gwern17 Jun 20263m
“Predicting LLM Safety Before Release by Simulating Deployment” by Tomek Korbak, Marcus Williams, micahcarroll, Cameron Raymond, Hannah Sheahan17 Jun 20262m
“What are some angles of attack for making continual learning safer?” by Rauno Arike, RohanS, Owen Terry, Achu Menon, Zhijing Jin, Francis Rhys Ward, Seth Herd16 Jun 202625m
“Synthetic document finetuning for instilling positive traits” by CallumMcDougall, Arthur Conmy, Neel Nanda16 Jun 202618m
“How might continual learning affect safety and alignment?” by Rauno Arike, RohanS, Owen Terry, Achu Menon, Zhijing Jin, Francis Rhys Ward, Seth Herd13 Jun 202630m
“SFT Drives Gemini’s Safety Properties” by Josh Engels, Arthur Conmy, bilalchughtai, Neel Nanda13 Jun 20263m