Metraipolis › archive › dc_2d3258f30ff3
Fusion Training for Mathematical Generalization in Large Language Models — arXiv:2608.09893
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.09893
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T08:50:32.125Z
- SHA-256 of the archived text
eba2a1b8d03e9cd4681c82b2336c1d85b5c916909430dd33a7d06300a1cccdc8- Stable URL of this page
https://metraipolis.com/documento/dc_2d3258f30ff3
How to cite
Fusion Training for Mathematical Generalization in Large Language Models — arXiv:2608.09893 — arXiv (en), https://arxiv.org/abs/2608.09893 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2d3258f30ff3 (Metraipolis archive dc_2d3258f30ff3, sha256 eba2a1b8d03e9cd4681c82b2336c1d85b5c916909430dd33a7d06300a1cccdc8)
BibTeX
@misc{dc2d3258f30ff3,
title = {Fusion Training for Mathematical Generalization in Large Language Models — arXiv:2608.09893},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.09893},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2d3258f30ff3; sha256 eba2a1b8d03e9cd4681c82b2336c1d85b5c916909430dd33a7d06300a1cccdc8},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.09893. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Thinking Mode Fusion (TMF) enables large language models to support both concise responses and long-form reasoning by unifying a non-thinking mode and a thinking mode within a single model. However, its training dynamics, including the \emph{data ratio} and \emph{training schedule} between the two modes, remain underexplored. In this work, we present a systematic study of TMF by analyzing the effects of the training schedule and data ratio between thinking and non-thinking modes. Focusing on mathematical problem solving, we construct a benchmark with multiple thinking-to-non-thinking data ratios and three training schedules. Our results reveal an asymmetric interaction between the two modes: increasing the ratio of non-thinking supervision reduces the accuracy of the thinking mode. We further show that different training schedules modulate this trade-off and that the optimal schedule depends on the data ratio. Finally, we quantify a negative correlation between non-thinking and thinking mode supervision, highlighting an inherent tension between these two modes. These findings provide practical guidance for designing effective TMF training settings. All code and data are released to support further research at: \href{ this https URL }{\textbf{Fusion Bench}}.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-11 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.09893 — 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 2026-10-19T08:49:41.947Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).