Agent psychometrics: Task-level performance prediction in agentic coding benchmarks
arXiv:2604.00594v1 Announce Type: new Abstract: As the focus in LLM-based coding shifts from static single-step code generation to multi-step agentic interaction with tools and environments, …
Chris Ge, Daria Kryvosheieva, Daniel Fried, Uzay Girit, Kaivalya Hariharan
2 views
