More Programs or More Rolls? Separating Coverage from Specialization in LLM Harnesses
arXiv:2609.35873v1 Announce Type: new Abstract: Automated generation of LLM harnesses promises to improve inference through task specialization. Yet additional answer coverage can arise from repeated execution of the same program, making...
Read more: https://arxiv.org/abs/2609.35873
Scan with your phone
Scan with your phone