Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2505.24319 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908385256931328 |
|---|---|
| author | Shi, Yuntao Luo, Yi Gong, Yeyun Lin, Chen |
| author_facet | Shi, Yuntao Luo, Yi Gong, Yeyun Lin, Chen |
| contents | Large Language Models (LLMs) have achieved remarkable success in various domains. However, when handling long-form text modification tasks, they still face two major problems: (1) producing undesired modifications by inappropriately altering or summarizing irrelevant content, and (2) missing necessary modifications to implicitly related passages that are crucial for maintaining document coherence. To address these issues, we propose HiCaM, a Hierarchical-Causal Modification framework that operates through a hierarchical summary tree and a causal graph. Furthermore, to evaluate HiCaM, we derive a multi-domain dataset from various benchmarks, providing a resource for assessing its effectiveness. Comprehensive evaluations on the dataset demonstrate significant improvements over strong LLMs, with our method achieving up to a 79.50\% win rate. These results highlight the comprehensiveness of our approach, showing consistent performance improvements across multiple models and domains. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_24319 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | HiCaM: A Hierarchical-Causal Modification Framework for Long-Form Text Modification Shi, Yuntao Luo, Yi Gong, Yeyun Lin, Chen Computation and Language Large Language Models (LLMs) have achieved remarkable success in various domains. However, when handling long-form text modification tasks, they still face two major problems: (1) producing undesired modifications by inappropriately altering or summarizing irrelevant content, and (2) missing necessary modifications to implicitly related passages that are crucial for maintaining document coherence. To address these issues, we propose HiCaM, a Hierarchical-Causal Modification framework that operates through a hierarchical summary tree and a causal graph. Furthermore, to evaluate HiCaM, we derive a multi-domain dataset from various benchmarks, providing a resource for assessing its effectiveness. Comprehensive evaluations on the dataset demonstrate significant improvements over strong LLMs, with our method achieving up to a 79.50\% win rate. These results highlight the comprehensiveness of our approach, showing consistent performance improvements across multiple models and domains. |
| title | HiCaM: A Hierarchical-Causal Modification Framework for Long-Form Text Modification |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2505.24319 |