Smoothie-Qwen: Post-Hoc Smoothing to Reduce Language Bias in Multilingual LLMs

• URL: http://arxiv.org/abs/2507.05686v1
• Date: Tue, 08 Jul 2025 05:30:51 GMT
• Title: Smoothie-Qwen: Post-Hoc Smoothing to Reduce Language Bias in Multilingual LLMs
• Authors: SeungWon Ji, Jungyup Lee, Jemin Kim, Sang Park, SeungJae Lee,
• Abstract summary: Smoothie-Qwen is a lightweight, post-hoc method that mitigates language bias without retraining.<n>applied to the Qwen model, our method reduces unintended Chinese output by over 95%.
• Score: 4.881694369042022
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Multilingual large language models (LLMs) often exhibit language confusion, a tendency to generate responses in a dominant language irrespective of the prompt's language. To address this, we propose Smoothie-Qwen, a lightweight, post-hoc method that mitigates language bias without retraining. This technique selectively adjusts token-level output probabilities to effectively suppress undesired language generation. Applied to the Qwen model, our method reduces unintended Chinese output by over 95% while preserving task accuracy on multilingual benchmarks. This work provides a practical and efficient solution for enhancing the language controllability of LLMs, making them more reliable for global applications.

Related papers

• CausalAbstain: Enhancing Multilingual LLMs with Causal Reasoning for Trustworthy Abstention [9.76878200328024]
  Large Language Models (LLMs) often exhibit knowledge disparities across languages.<n>We introduce textitCausalAbstain, a method that helps LLMs determine whether to utilize multiple generated feedback responses.<n>Experiments demonstrate that textitCausalAbstain effectively selects helpful feedback and enhances abstention decisions with interpretability.
  arXiv  Detail & Related papers  (2025-05-31T11:35:31Z)
• Controlling Language Confusion in Multilingual LLMs [0.0]
  Large language models often suffer from language confusion, a phenomenon where responses are partially or entirely generated in unintended languages.<n>We find that ORPO, which adds penalties for unwanted output styles to standard SFT, effectively suppresses language-confused generations even at high decoding temperatures.<n>Our findings suggest that incorporating appropriate penalty terms can mitigate language confusion in low-resource settings with limited data.
  arXiv  Detail & Related papers  (2025-05-25T12:15:31Z)
• Understanding and Mitigating Language Confusion in LLMs [76.96033035093204]
  We evaluate 15 typologically diverse languages with existing and newly-created English and multilingual prompts.<n>We find that Llama Instruct and Mistral models exhibit high degrees of language confusion.<n>We find that language confusion can be partially mitigated via few-shot prompting, multilingual SFT and preference tuning.
  arXiv  Detail & Related papers  (2024-06-28T17:03:51Z)
• No Train but Gain: Language Arithmetic for training-free Language Adapters enhancement [59.37775534633868]
  We introduce a novel method called language arithmetic, which enables training-free post-processing. The effectiveness of the proposed solution is demonstrated on three downstream tasks in a MAD-X-based set of cross-lingual schemes.
  arXiv  Detail & Related papers  (2024-04-24T08:52:40Z)
• The Ups and Downs of Large Language Model Inference with Vocabulary Trimming by Language Heuristics [74.99898531299148]
  This research examines vocabulary trimming (VT) inspired by restricting embedding entries to the language of interest to bolster time and memory efficiency. We apply two languages to trim the full vocabulary - Unicode-based script filtering and corpus-based selection - to different language families and sizes. It is found that VT reduces the memory usage of small models by nearly 50% and has an upper bound of 25% improvement in generation speed.
  arXiv  Detail & Related papers  (2023-11-16T09:35:50Z)
• PLUG: Leveraging Pivot Language in Cross-Lingual Instruction Tuning [46.153828074152436]
  We propose a pivot language guided generation approach to enhance instruction tuning in lower-resource languages. It trains the model to first process instructions in the pivot language, and then produce responses in the target language. Our approach demonstrates a significant improvement in the instruction-following abilities of LLMs by 29% on average.
  arXiv  Detail & Related papers  (2023-11-15T05:28:07Z)
• Democratizing LLMs for Low-Resource Languages by Leveraging their English Dominant Abilities with Linguistically-Diverse Prompts [75.33019401706188]
  Large language models (LLMs) are known to effectively perform tasks by simply observing few exemplars. We propose to assemble synthetic exemplars from a diverse set of high-resource languages to prompt the LLMs to translate from any language into English. Our unsupervised prompting method performs on par with supervised few-shot learning in LLMs of different sizes for translations between English and 13 Indic and 21 African low-resource languages.
  arXiv  Detail & Related papers  (2023-06-20T08:27:47Z)
• Soft Language Clustering for Multilingual Model Pre-training [57.18058739931463]
  We propose XLM-P, which contextually retrieves prompts as flexible guidance for encoding instances conditionally. Our XLM-P enables (1) lightweight modeling of language-invariant and language-specific knowledge across languages, and (2) easy integration with other multilingual pre-training methods.
  arXiv  Detail & Related papers  (2023-06-13T08:08:08Z)
• InstructAlign: High-and-Low Resource Language Alignment via Continual Crosslingual Instruction Tuning [66.31509106146605]
  Large language models (LLMs) that are tuned with instructions have demonstrated remarkable capabilities in various tasks and languages. However, their ability to generalize to underrepresented languages is limited due to the scarcity of available data. We propose InstructAlign which uses continual crosslingual instruction tuning to enable LLMs to align new unseen languages with previously learned high-resource languages.
  arXiv  Detail & Related papers  (2023-05-23T02:51:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.