Metraipolis › archive › dc_562f41392f0e
Train What You Deploy:Token-Faithful Post-Training of a Production Coding — arXiv:2609.04678
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.04678
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T22:20:49.474Z
- SHA-256 of the archived text
53b0af63a3b85227ee42c168082bdfe41b08629e7005f5299fae81d8abd27b95- Stable URL of this page
https://metraipolis.com/documento/dc_562f41392f0e
How to cite
Train What You Deploy:Token-Faithful Post-Training of a Production Coding — arXiv:2609.04678 — arXiv (en), https://arxiv.org/abs/2609.04678 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_562f41392f0e (Metraipolis archive dc_562f41392f0e, sha256 53b0af63a3b85227ee42c168082bdfe41b08629e7005f5299fae81d8abd27b95)
BibTeX
@misc{dc562f41392f0e,
title = {Train What You Deploy:Token-Faithful Post-Training of a Production Coding — arXiv:2609.04678},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.04678},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_562f41392f0e; sha256 53b0af63a3b85227ee42c168082bdfe41b08629e7005f5299fae81d8abd27b95},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.04678. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Existing post-training pipelines for coding and terminal agents suffer severe token and control fidelity errors: simplified training environments mismatch production deployments, and offline token reconstruction from agent logs distorts original prompts and conflates policy calls with background model operations. We present a fidelity-aware training coupling framework that retains trainer-side sampling over original prompts, eliminates spurious model calls via a negotiated training protocol, and restricts loss computation to verifiable token spans with closed-failure guarantees. We further propose Certified Divergence Proximal Policy Optimization (C-DPPO), which establishes tight two-sided TV certification bounds, adaptive-K rules, budget-aware sequence guarantees, and error-robust policy masking atop standard DPPO. Evaluated on matched Baize5B and Baize10B models with identical training and test protocols on TMax-100, C-DPPO yields a consistent +3.0-point performance gain over standard DPPO across model scales. Certificate audits validate the reliability and full operational coverage of our certified training pipeline. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-07 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.04678 — 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 2026-10-16T22:18:11.376Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).