We have 4 of 13 papers
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
This suite provides a new efficient lens on generalization: select intermediate pre-training checkpoints that strongly generalize reasoning and alignment, better than the final pre- or mid-training checkpoints, and select pre-training data that controls and stabilizes generalization dynamics.
A new benchmark to more comprehensively evaluate CUAs'misuse risks, CUAHarm, and explores using LMs to monitor CUAs'actions, finding monitoring unsafe computer-using actions is significantly harder than monitoring conventional unsafe chatbot responses.
A benchmark score is a measurement instrument, yet most benchmarks read each item at a single canonical surface form. We ask whether that reading is faithful: when an item's intent is held fixed and only its meaning-preserving surface form varies, does the canonical-form score estimate model behavior well, and how much...
Using human ideas as the AARs' initial research direction does not improve performance, suggesting current AARs may not need guidance from experienced researchers, and suggests that automating alignment research on well-characterized failures may be practical in the near term.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
