Metraipolis › archive › dc_02a393a85aae
Reasoning-Trace Overflow in Code Generation: An Empirical Study of Explicit Reasoning Models — arXiv:2611.06095
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.06095
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-05T18:16:55.017Z
- SHA-256 of the archived text
22c46b98d61b90b6a53106b81dbf171a5a7583e0ac9af8d0eea0f4fda52bf162- Stable URL of this page
https://metraipolis.com/documento/dc_02a393a85aae
How to cite
Reasoning-Trace Overflow in Code Generation: An Empirical Study of Explicit Reasoning Models — arXiv:2611.06095 — arXiv (en), https://arxiv.org/abs/2611.06095 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_02a393a85aae (Metraipolis archive dc_02a393a85aae, sha256 22c46b98d61b90b6a53106b81dbf171a5a7583e0ac9af8d0eea0f4fda52bf162)
BibTeX
@misc{dc02a393a85aae,
title = {Reasoning-Trace Overflow in Code Generation: An Empirical Study of Explicit Reasoning Models — arXiv:2611.06095},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.06095},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_02a393a85aae; sha256 22c46b98d61b90b6a53106b81dbf171a5a7583e0ac9af8d0eea0f4fda52bf162},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.06095. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Explicit reasoning can help language models generate code, but it also introduces a practical risk: the model may spend so much of its output budget reasoning that it never produces usable code. We study this behaviour in relatively small open-weight reasoning models, framing it as reasoning-trace overflow. Across seven models and six code benchmark settings, we measure how much of a fixed 32k-token output budget is spent on reasoning, whether that reasoning improves pass@1, and whether lightweight inference-time controls can mitigate overflow. Our results show that the value of reasoning is highly conditional. Overflow is rare on simpler function-level tasks, but becomes common on harder code-generation benchmarks and for particular model families: on several model--benchmark pairs, more than half of generations exhaust the full 32k-token budget while still reasoning. Most complete overflow is associated with repetitive reasoning loops, although substantial long reasoning also occurs without looping. Disabling reasoning can therefore be competitive for overflow-prone models, simply because it ensures that an answer is produced. At the same time, reasoning remains valuable for stronger models on harder algorithmic and code-understanding tasks. Finally, prompt steering can substantially reduce overflow for some models but is inconsistent, while budgeted transitions provide more direct control and can improve both performance and token efficiency for overflow-prone models. These findings suggest that explicit reasoning should be treated as a controllable resource for code generation, rather than an always-on capability. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI) ; Software Engineering (cs.SE). Pagina del lavoro: https://arxiv.org/abs/2611.06095 — 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 2026-11-05T18:12:49.005Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).