Metraipolis › archive › dc_ed1b95c5ceeb

Astar: Learning to Propose Evolution Directions for Self-Evolving Industrial AI Systems — arXiv:2608.27287

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.27287
License (as archived): CC0-1.0
Archived: 2026-10-17T17:26:38.299Z
SHA-256 of the archived text: d5f37fefc39ce90288d380474f003f24b1db3b762a5bb3d99d2252d659a5223a
Stable URL of this page: https://metraipolis.com/documento/dc_ed1b95c5ceeb

How to cite

Astar: Learning to Propose Evolution Directions for Self-Evolving Industrial AI Systems — arXiv:2608.27287 — arXiv (en), https://arxiv.org/abs/2608.27287 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ed1b95c5ceeb (Metraipolis archive dc_ed1b95c5ceeb, sha256 d5f37fefc39ce90288d380474f003f24b1db3b762a5bb3d99d2252d659a5223a)

BibTeX

@misc{dced1b95c5ceeb,
  title = {Astar: Learning to Propose Evolution Directions for Self-Evolving Industrial AI Systems — arXiv:2608.27287},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.27287},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ed1b95c5ceeb; sha256 d5f37fefc39ce90288d380474f003f24b1db3b762a5bb3d99d2252d659a5223a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.27287. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Modern AI systems advance through continuous iteration: a loop of proposing evolution directions, implementing code, training, and evaluation. While the latter three stages are increasingly automated, the starting point --- proposing effective evolution directions --- remains a critical bottleneck that still relies heavily on senior experts. In this work, we explore whether AI can take over this role. We find that general-purpose LLMs, even the advanced GPT-5.5, offer only generic and misaligned suggestions: the required expertise is accumulated through experience rather than explicitly codified, and thus hard to inject directly. To this end, we propose Astar, a training-based approach that learns a specialized evolution-guiding model from the abundant iteration histories of industrial systems. Realizing this idea, however, raises four challenges: sparse supervision, noisy data, a vast direction space, and prohibitively expensive verification. We address them along two fronts. On the data side, we design a pipeline that turns noisy historical commits into a large, clean evolutionary corpus via pairwise sample expansion and noise filtering. On the model side, we train the model through mid-training, SFT, and RL, guiding evolution direction generation with hierarchical hints and using the reward model in RL as a fast surrogate evaluator. Astar has been deployed in Alibaba's Lazada advertising system for evolution direction proposal. Astar-8B achieves a single-proposal success rate of 0.6786 in real-execution evaluation, far exceeding human experts (0.3229) and the strongest general-purpose LLM (0.3071). More importantly, Astar closes the loop and enables fully automatic iteration: it guided 20 consecutive iterations over two weeks, improving offline Hitrate@200 by 23.6%, while an online A/B test yielded relative lifts of 4.86% in GMV and 1.82% in advertising revenue.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-28 in cs.IR).
Categorie arXiv: Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2608.27287 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T17:18:28.088Z dalla pagina https://arxiv.org/catchup/cs.IR/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).