SOAEsV2-7B/72B: Full-Pipeline Optimization for State-Owned Enterprise LLMs via Continual Pre-Training, Domain-Progressive SFT and Distillation-Enhanced Speculative Decoding
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918012971384832 |
|---|---|
| author | Deng, Jingyang Chen, Ran Cheng, Jo-Ku Ma, Jinwen |
| author_facet | Deng, Jingyang Chen, Ran Cheng, Jo-Ku Ma, Jinwen |
| contents | This study addresses key challenges in developing domain-specific large language models (LLMs) for Chinese state-owned assets and enterprises (SOAEs), where current approaches face three limitations: 1) constrained model capacity that limits knowledge integration and cross-task adaptability; 2) excessive reliance on domain-specific supervised fine-tuning (SFT) data, which neglects the broader applicability of general language patterns; and 3) inefficient inference acceleration for large models processing long contexts. In this work, we propose SOAEsV2-7B/72B, a specialized LLM series developed via a three-phase framework: 1) continual pre-training integrates domain knowledge while retaining base capabilities; 2) domain-progressive SFT employs curriculum-based learning strategy, transitioning from weakly relevant conversational data to expert-annotated SOAEs datasets to optimize domain-specific tasks; 3) distillation-enhanced speculative decoding accelerates inference via logit distillation between 72B target and 7B draft models, achieving 1.39-1.52$\times$ speedup without quality loss. Experimental results demonstrate that our domain-specific pre-training phase maintains 99.8% of original general language capabilities while significantly improving domain performance, resulting in a 1.08$\times$ improvement in Rouge-1 score and a 1.17$\times$ enhancement in BLEU-4 score. Ablation studies further show that domain-progressive SFT outperforms single-stage training, achieving 1.02$\times$ improvement in Rouge-1 and 1.06$\times$ in BLEU-4. Our work introduces a comprehensive, full-pipeline approach for optimizing SOAEs LLMs, bridging the gap between general language capabilities and domain-specific expertise. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_04723 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SOAEsV2-7B/72B: Full-Pipeline Optimization for State-Owned Enterprise LLMs via Continual Pre-Training, Domain-Progressive SFT and Distillation-Enhanced Speculative Decoding Deng, Jingyang Chen, Ran Cheng, Jo-Ku Ma, Jinwen Computation and Language This study addresses key challenges in developing domain-specific large language models (LLMs) for Chinese state-owned assets and enterprises (SOAEs), where current approaches face three limitations: 1) constrained model capacity that limits knowledge integration and cross-task adaptability; 2) excessive reliance on domain-specific supervised fine-tuning (SFT) data, which neglects the broader applicability of general language patterns; and 3) inefficient inference acceleration for large models processing long contexts. In this work, we propose SOAEsV2-7B/72B, a specialized LLM series developed via a three-phase framework: 1) continual pre-training integrates domain knowledge while retaining base capabilities; 2) domain-progressive SFT employs curriculum-based learning strategy, transitioning from weakly relevant conversational data to expert-annotated SOAEs datasets to optimize domain-specific tasks; 3) distillation-enhanced speculative decoding accelerates inference via logit distillation between 72B target and 7B draft models, achieving 1.39-1.52$\times$ speedup without quality loss. Experimental results demonstrate that our domain-specific pre-training phase maintains 99.8% of original general language capabilities while significantly improving domain performance, resulting in a 1.08$\times$ improvement in Rouge-1 score and a 1.17$\times$ enhancement in BLEU-4 score. Ablation studies further show that domain-progressive SFT outperforms single-stage training, achieving 1.02$\times$ improvement in Rouge-1 and 1.06$\times$ in BLEU-4. Our work introduces a comprehensive, full-pipeline approach for optimizing SOAEs LLMs, bridging the gap between general language capabilities and domain-specific expertise. |
| title | SOAEsV2-7B/72B: Full-Pipeline Optimization for State-Owned Enterprise LLMs via Continual Pre-Training, Domain-Progressive SFT and Distillation-Enhanced Speculative Decoding |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2505.04723 |