Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.12559 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911592106426368 |
|---|---|
| author | Wang, Peng Zhou, Biyu Tang, Xuehai Han, Jizhong Hu, Songlin |
| author_facet | Wang, Peng Zhou, Biyu Tang, Xuehai Han, Jizhong Hu, Songlin |
| contents | Unstructured model editing aims to update models with real-world text, yet existing methods often memorize text holistically without reliable fine-grained fact access. To address this, we propose FABLE, a hierarchical framework that decouples fine-grained fact injection from holistic text generation. FABLE follows a two-stage, fact-first strategy: discrete facts are anchored in shallow layers, followed by minimal updates to deeper layers to produce coherent text. This decoupling resolves the mismatch between holistic recall and fine-grained fact access, reflecting the unidirectional Transformer flow in which surface-form generation amplifies rather than corrects underlying fact representations. We also introduce UnFine, a diagnostic benchmark with fine-grained question-answer pairs and fact-level metrics for systematic evaluation. Experiments show that FABLE substantially improves fine-grained question answering while maintaining state-of-the-art holistic editing performance. Our code is publicly available at https://github.com/caskcsg/FABLE. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_12559 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | FABLE: Fine-grained Fact Anchoring for Unstructured Model Editing Wang, Peng Zhou, Biyu Tang, Xuehai Han, Jizhong Hu, Songlin Computation and Language Unstructured model editing aims to update models with real-world text, yet existing methods often memorize text holistically without reliable fine-grained fact access. To address this, we propose FABLE, a hierarchical framework that decouples fine-grained fact injection from holistic text generation. FABLE follows a two-stage, fact-first strategy: discrete facts are anchored in shallow layers, followed by minimal updates to deeper layers to produce coherent text. This decoupling resolves the mismatch between holistic recall and fine-grained fact access, reflecting the unidirectional Transformer flow in which surface-form generation amplifies rather than corrects underlying fact representations. We also introduce UnFine, a diagnostic benchmark with fine-grained question-answer pairs and fact-level metrics for systematic evaluation. Experiments show that FABLE substantially improves fine-grained question answering while maintaining state-of-the-art holistic editing performance. Our code is publicly available at https://github.com/caskcsg/FABLE. |
| title | FABLE: Fine-grained Fact Anchoring for Unstructured Model Editing |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2604.12559 |