{"slug": "learning-to-translate-from-soft-to-hard-llm-prompts", "title": "Learning to Translate from Soft to Hard LLM Prompts", "summary": "Researchers developed a translation model that converts soft prompts—parameter-efficient but uninterpretable LLM adaptations—into natural language text. The translated prompts outperformed existing interpretability methods like InSPEcT on multiple datasets, producing more fluent and accurate verbalizations. This approach enables soft prompts optimized on small open-source models to be deployed as portable text prompts on larger closed-API models, where they can exceed the original soft prompt's performance and even rival few-shot learning.", "body_md": "arXiv:2605.27642v1 Announce Type: new\nAbstract: Soft prompt tuning is a parameter-efficient method for adapting LLMs to specific tasks, but suffers from a lack of interpretability. Building on recent work on interpreting soft prompts (Ramati et al., 2024), we explore how training a dedicated soft prompt to natural language translation model can yield higher translation quality. In particular, in both quantitative and qualitative comparisons on multiple Datasets of Datasets (DoDs), we demonstrate that our translator produces fluent, accurate verbalizations that outperforms existing training-free methods like InSPEcT. In addition to advancing interpretability, our work suggests a promising downstream application: soft prompts optimized on small, open-source models can be translated into portable text prompts that, when deployed on larger closed-API models, exceed the performance of the original soft prompt and, in some cases, even few-shot learning.", "url": "https://wpnews.pro/news/learning-to-translate-from-soft-to-hard-llm-prompts", "canonical_source": "https://arxiv.org/abs/2605.27642", "published_at": "2026-05-28 04:00:00+00:00", "updated_at": "2026-05-28 04:35:34.086109+00:00", "lang": "en", "topics": ["large-language-models", "natural-language-processing", "machine-learning", "artificial-intelligence", "ai-research"], "entities": ["Ramati et al.", "InSPEcT"], "alternates": {"html": "https://wpnews.pro/news/learning-to-translate-from-soft-to-hard-llm-prompts", "markdown": "https://wpnews.pro/news/learning-to-translate-from-soft-to-hard-llm-prompts.md", "text": "https://wpnews.pro/news/learning-to-translate-from-soft-to-hard-llm-prompts.txt", "jsonld": "https://wpnews.pro/news/learning-to-translate-from-soft-to-hard-llm-prompts.jsonld"}}