Models from "Efficient Reasoning Training Does Not Always Harm CoT Faithfulness and Monitorability".
Samuel Lewis-Lim PRO
Samll
AI & ML interests
None yet
Recent Activity
upvoted a paper 1 day ago
Synthetic Pre-pretraining Survives Scale, but Not as a Grammatical Prior updated a collection 1 day ago
efficient-reasoning updated a collection 1 day ago
efficient-reasoningOrganizations
None yet