Viewing profile — serendip-ml

Interesting question is if the CoE Audit evaluates evidence chains recursively (each cited paper's own evidence chain), or stops at citation existence + method-code alignment for t…

Author here. Curious what others think about the verb-based approach vs pure schema validation. Happy to discuss trade-offs.

Hi HN, Karpathy's recent post [1] described Claws as "a new layer on top of LLM agents, taking orchestration, scheduling, context, tool calls to a next level." That's the right fra…

The compression analogy is interesting. Another way of looking at it could be fine-tuning as "knowing what to leave out" - a 3B model for example tuned for a narrow task doesn't ne…