HyperEdit: Unlocking Instruction-based Text Editing in LLMs via Hypernetworks
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918248385085440 |
|---|---|
| author | Zeng, Yiming Cao, Jinghan Li, Zexin Yu, Wanhao Ye, Zhankai Xiang, Dawei Hua, Ting Liu, Xin Gao, Shangqian Yu, Tingting |
| author_facet | Zeng, Yiming Cao, Jinghan Li, Zexin Yu, Wanhao Ye, Zhankai Xiang, Dawei Hua, Ting Liu, Xin Gao, Shangqian Yu, Tingting |
| contents | Instruction-based text editing is increasingly critical for real-world applications such as code editors (e.g., Cursor), but Large Language Models (LLMs) continue to struggle with this task. Unlike free-form generation, editing requires faithfully implementing user instructions while preserving unchanged content, as even minor unintended modifications can break functionality. Existing approaches treat editing as generic text generation, leading to two key failures: they struggle to faithfully align edits with diverse user intents, and they often over-edit unchanged regions. We propose HyperEdit to address both issues. First, we introduce hypernetwork-based dynamic adaptation that generates request-specific parameters, enabling the model to tailor its editing strategy to each instruction. Second, we develop difference-aware regularization that focuses supervision on modified spans, preventing over-editing while ensuring precise, minimal changes. HyperEdit achieves a 9%--30% relative improvement in BLEU on modified regions over state-of-the-art baselines, despite utilizing only 3B parameters. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_12544 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | HyperEdit: Unlocking Instruction-based Text Editing in LLMs via Hypernetworks Zeng, Yiming Cao, Jinghan Li, Zexin Yu, Wanhao Ye, Zhankai Xiang, Dawei Hua, Ting Liu, Xin Gao, Shangqian Yu, Tingting Computation and Language Machine Learning Instruction-based text editing is increasingly critical for real-world applications such as code editors (e.g., Cursor), but Large Language Models (LLMs) continue to struggle with this task. Unlike free-form generation, editing requires faithfully implementing user instructions while preserving unchanged content, as even minor unintended modifications can break functionality. Existing approaches treat editing as generic text generation, leading to two key failures: they struggle to faithfully align edits with diverse user intents, and they often over-edit unchanged regions. We propose HyperEdit to address both issues. First, we introduce hypernetwork-based dynamic adaptation that generates request-specific parameters, enabling the model to tailor its editing strategy to each instruction. Second, we develop difference-aware regularization that focuses supervision on modified spans, preventing over-editing while ensuring precise, minimal changes. HyperEdit achieves a 9%--30% relative improvement in BLEU on modified regions over state-of-the-art baselines, despite utilizing only 3B parameters. |
| title | HyperEdit: Unlocking Instruction-based Text Editing in LLMs via Hypernetworks |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2512.12544 |