Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2508.02452 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916879803613184 |
|---|---|
| author | Bystroński, Mateusz Piotrowski, Grzegorz Chawla, Nitesh V. Kajdanowicz, Tomasz |
| author_facet | Bystroński, Mateusz Piotrowski, Grzegorz Chawla, Nitesh V. Kajdanowicz, Tomasz |
| contents | Recent advances have shown that optimizing prompts for Large Language Models (LLMs) can significantly improve task performance, yet many optimization techniques rely on heuristics or manual exploration. We present LatentPrompt, a model-agnostic framework for prompt optimization that leverages latent semantic space to automatically generate, evaluate, and refine candidate prompts without requiring hand-crafted rules. Beginning with a set of seed prompts, our method embeds them in a continuous latent space and systematically explores this space to identify prompts that maximize task-specific performance. In a proof-of-concept study on the Financial PhraseBank sentiment classification benchmark, LatentPrompt increased classification accuracy by approximately 3 percent after a single optimization cycle. The framework is broadly applicable, requiring only black-box access to an LLM and an automatic evaluation metric, making it suitable for diverse domains and tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_02452 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LatentPrompt: Optimizing Promts in Latent Space Bystroński, Mateusz Piotrowski, Grzegorz Chawla, Nitesh V. Kajdanowicz, Tomasz Computation and Language Recent advances have shown that optimizing prompts for Large Language Models (LLMs) can significantly improve task performance, yet many optimization techniques rely on heuristics or manual exploration. We present LatentPrompt, a model-agnostic framework for prompt optimization that leverages latent semantic space to automatically generate, evaluate, and refine candidate prompts without requiring hand-crafted rules. Beginning with a set of seed prompts, our method embeds them in a continuous latent space and systematically explores this space to identify prompts that maximize task-specific performance. In a proof-of-concept study on the Financial PhraseBank sentiment classification benchmark, LatentPrompt increased classification accuracy by approximately 3 percent after a single optimization cycle. The framework is broadly applicable, requiring only black-box access to an LLM and an automatic evaluation metric, making it suitable for diverse domains and tasks. |
| title | LatentPrompt: Optimizing Promts in Latent Space |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2508.02452 |