Metraipolis › archive › dc_bd63238b59a2
What is Missing from AI Post-Training AI: An Empirical Analysis — arXiv:2608.19072
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.19072
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T13:44:42.274Z
- SHA-256 of the archived text
474bdf60bf17bf25304fcef7716561ae6ce704ef2893bf0ddf8ebc952fd8c620- Stable URL of this page
https://metraipolis.com/documento/dc_bd63238b59a2
How to cite
What is Missing from AI Post-Training AI: An Empirical Analysis — arXiv:2608.19072 — arXiv (en), https://arxiv.org/abs/2608.19072 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bd63238b59a2 (Metraipolis archive dc_bd63238b59a2, sha256 474bdf60bf17bf25304fcef7716561ae6ce704ef2893bf0ddf8ebc952fd8c620)
BibTeX
@misc{dcbd63238b59a2,
title = {What is Missing from AI Post-Training AI: An Empirical Analysis — arXiv:2608.19072},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.19072},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bd63238b59a2; sha256 474bdf60bf17bf25304fcef7716561ae6ce704ef2893bf0ddf8ebc952fd8c620},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.19072. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language model (LLM) agents can now post-train an LLM end-to-end, raising the prospect of recursive self-improvement (RSI). Yet this progress is measured by aggregate benchmark scores, which cannot tell whether an agent executes a fixed plan well or strategically revises the plan when it fails. We separate these two capabilities: execution-level capability, iterating within an established training strategy, and strategy-level capability, revising that strategy as experimental evidence accumulates. Analyzing 1,338 post-training trajectories of frontier agents, we find that agents reliably execute post-training but lock into a default strategy, which follows the agent rather than the task, and only 2.1% of transitions between adjacent training runs ever change strategy. We then test whether the agent lacks experience, reasoning, or the decision to switch. (1) Experience improves execution but not the strategy. (2) Additional reasoning compute yields front-loaded gains on easier tasks but refines, rather than revises, the committed strategy. (3) Human review before training changes which strategy the agent locks into, not whether it locks in, whereas a single mid-run instruction outperforms the agent's own continuation by up to 17.44 points under the same budget. In conclusion, what the agent lacks is the decision to reopen a committed strategy and try another one. Realizing RSI therefore calls for interaction protocols and training signals that make strategy revision an explicit, rewarded decision. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-20 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.19072 — 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 2026-10-18T13:41:17.890Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-20?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).