Metraipolis › archive › dc_115f43382cac

Say Less, Mean Better: Training-Free Verbosity Control for Language Models — arXiv:2611.04107

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.04107
License (as archived)
CC0-1.0
Archived
2026-11-04T12:32:51.085Z
SHA-256 of the archived text
6698569e4d56146881433cad6991c84ace3967bd2af3033f77d851b0739b5b5a
Stable URL of this page
https://metraipolis.com/documento/dc_115f43382cac

How to cite

Say Less, Mean Better: Training-Free Verbosity Control for Language Models — arXiv:2611.04107 — arXiv (en), https://arxiv.org/abs/2611.04107 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_115f43382cac (Metraipolis archive dc_115f43382cac, sha256 6698569e4d56146881433cad6991c84ace3967bd2af3033f77d851b0739b5b5a)
BibTeX
@misc{dc115f43382cac,
  title = {Say Less, Mean Better: Training-Free Verbosity Control for Language Models — arXiv:2611.04107},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04107},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_115f43382cac; sha256 6698569e4d56146881433cad6991c84ace3967bd2af3033f77d851b0739b5b5a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04107. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models (LLMs) often generate responses longer than necessary, increasing inference latency and computational cost. While prompting can encourage conciseness, controlling output length without compromising response completeness remains challenging. Aggressive early stopping can reduce generation costs but risks terminating responses prematurely. We introduce MinimalLM, a training-free framework for reliable verbosity control that combines uncertainty-aware, length-dependent stop-token adjustment with structural completeness guardrails. MinimalLM separates the decision of when to stop from whether stopping is structurally appropriate, coordinating response-level instructions, decoding-time intervention, and completion checks without modifying model parameters or requiring auxiliary models. We evaluate MinimalLM across three open-source LLMs and 330 questions, producing 5,940 responses across baseline and ablation configurations. MinimalLM reduces generated tokens by 47.6%, accelerates individual responses by 19-68%, and eliminates structurally truncated outputs in the primary evaluation. Despite substantial token savings, it achieves an average quality score of 8.04/10, compared with 8.42/10 for unmodified models and 6.46/10 for length-matched truncation. Evaluations across diverse subject domains (such as healthcare, programming, law, and science) demonstrate robust efficiency gains, while ablations establish the complementary roles of decoding control and completeness constraints. Our findings highlight that efficient generation requires more than producing fewer tokens: it requires controlling when generation ends while preserving response integrity. MinimalLM offers a lightweight, training-free approach toward this goal.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2611.04107 — 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 2026-11-04T12:27:02.030Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).