Why Solve It Twice? Hierarchical Accumulation of Skills for Transfer-Efficient ML Engineering

Researchers introduced HASTE, a hierarchical multi-agent system that organizes cross-competition knowledge into three scope tiers to improve transfer efficiency in ML engineering. In tests on the MLE-Bench Lite benchmark, HASTE achieved a 77.3% medal rate using Claude Sonnet 4.6, with warm starts using 52% fewer refinement iterations. The system's tiered loading reached a 100% medal rate compared to 62.5% for flat loading, demonstrating that better knowledge organization can substitute for model strength and compute budget.

arXiv:2606.30911v1 Announce Type: new Abstract: ML engineering agents waste compute rediscovering known techniques because every competition is a cold start. We present HASTE, a hierarchical multi-agent system that organizes cross-competition knowledge into three scope tiers global, domain, and competition-specific , each coupled to a matching agent level. An orchestrator coordinates domain specialists and promotes learning between tiers via LLM-driven abstraction. A controlled ablation provides evidence for scoped loading: holding a 159-skill inventory constant across 8 competitions, tiered loading achieves a 100% medal rate while flat loading reaches only 62.5%, the same medal rate as loading no skills, and consumes 2x the output tokens. On the full MLE-Bench Lite benchmark 22 Kaggle competitions , HASTE reaches a medal rate of 77.3% using Claude Sonnet 4.6 at 12h per competition. In a cold-start run, the system begins with no accumulated skills. In warm-start runs, it reloads skills learned from earlier competitions, using only global and domain-level skills for transfer across competitions. Warm starts use 52% fewer refinement iterations, and the fraction of proposed changes kept by the agent rises from 42% at low inventory to 85% once 50+ skills are available. These results suggest that better knowledge organization can partly substitute for model strength and compute budget in ML-engineering agents.