Tag
The article argues that the output of AI coding agents is straining traditional CI systems, creating a bottleneck that cannot be fixed by hardware alone, and promotes Dagger as a smarter software stack solution.
This paper revisits Dataset Aggregation (DAgger) for training long-horizon LLM agents, demonstrating that turn-level teacher-student policy interpolation mitigates covariate shift and outperforms existing methods on software engineering benchmarks like SWE-bench Verified.