Metraipolis › archive › dc_d769661ea355
Mind the Cap: Output-Budget Regimes Change the Measured Multilingual Reasoning Gap — arXiv:2608.04160
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.04160
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T11:24:32.545Z
- SHA-256 of the archived text
20453da13a38f468947f29b9bc25329ef70b39d54311685f5e4d22f59098fe76- Stable URL of this page
https://metraipolis.com/documento/dc_d769661ea355
How to cite
Mind the Cap: Output-Budget Regimes Change the Measured Multilingual Reasoning Gap — arXiv:2608.04160 — arXiv (en), https://arxiv.org/abs/2608.04160 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d769661ea355 (Metraipolis archive dc_d769661ea355, sha256 20453da13a38f468947f29b9bc25329ef70b39d54311685f5e4d22f59098fe76)
BibTeX
@misc{dcd769661ea355,
title = {Mind the Cap: Output-Budget Regimes Change the Measured Multilingual Reasoning Gap — arXiv:2608.04160},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.04160},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d769661ea355; sha256 20453da13a38f468947f29b9bc25329ef70b39d54311685f5e4d22f59098fe76},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.04160. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multilingual evaluations report accuracy at a single output-token cap, but languages need different numbers of tokens to express the same content, so the cap is a hidden experimental variable. We test whether the native-vs-translate gap on MGSM (German, Thai, Swahili) is a token-budget artifact for Qwen3-8B and Llama-3.1-8B-Instruct under four prompting strategies. The measured gap swings by up to 57 points across budgets, length normalization moves it by up to 38.9 points where the cap binds, and at tight caps normalization can reverse which strategy scores higher. We prospectively froze the sweep's three Qwen peaks and its near-zero value at 1024 and evaluated them on 540,000 independently hard-capped decodes: a second frozen family of six Holm-corrected tests rejects every null. The frozen test at $B^*=1024$ still fails to reject because native accuracy has already saturated there; above saturation, the residual difference is a strategy-performance gap, not an identified reasoning deficit. The same truncation channel prices a cost-ordered adaptation ladder: a cross-fitted Thai vocabulary extension closes 0.0 points of the gap at the frozen budget and 4.9 points where 19% of traces still truncate. A third frozen family varies only the announced budget at a fixed enforced cap; announcing 128 rather than 2048 tokens moves Thai native accuracy by 5.1 points, so accuracy is not a function of the enforced cap alone. A correct-emission timing identity computed from one long-cap run matches the three pre-specified MGSM peaks to 0.65 points and, in an exploratory Qwen-only analysis of three further benchmarks, tracks held-out items to 0.92 points, locating the peak exactly in five of seven cells. Treat the output cap as an independent variable and report accuracy across the budget regime, not at a single budget. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-06 in cs.CL). Categorie arXiv: Computation and Language (cs.CL) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.04160 — 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 2026-10-31T11:13:23.153Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-06?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).