{"slug": "why-solve-it-twice-hierarchical-accumulation-of-skills-for-transfer-efficient-ml", "title": "Why Solve It Twice? Hierarchical Accumulation of Skills for Transfer-Efficient ML Engineering", "summary": "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.", "body_md": "arXiv:2606.30911v1 Announce Type: new\nAbstract: 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.", "url": "https://wpnews.pro/news/why-solve-it-twice-hierarchical-accumulation-of-skills-for-transfer-efficient-ml", "canonical_source": "https://arxiv.org/abs/2606.30911", "published_at": "2026-07-01 04:00:00+00:00", "updated_at": "2026-07-01 04:24:33.625286+00:00", "lang": "en", "topics": ["machine-learning", "ai-agents", "ai-research", "large-language-models", "mlops"], "entities": ["HASTE", "Claude Sonnet 4.6", "MLE-Bench Lite", "Kaggle"], "alternates": {"html": "https://wpnews.pro/news/why-solve-it-twice-hierarchical-accumulation-of-skills-for-transfer-efficient-ml", "markdown": "https://wpnews.pro/news/why-solve-it-twice-hierarchical-accumulation-of-skills-for-transfer-efficient-ml.md", "text": "https://wpnews.pro/news/why-solve-it-twice-hierarchical-accumulation-of-skills-for-transfer-efficient-ml.txt", "jsonld": "https://wpnews.pro/news/why-solve-it-twice-hierarchical-accumulation-of-skills-for-transfer-efficient-ml.jsonld"}}