Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2408.10642 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866929465884409856 |
|---|---|
| author | Xie, Shiming Chen, Hong Yu, Fred Sun, Zeye Wu, Xiuyu |
| author_facet | Xie, Shiming Chen, Hong Yu, Fred Sun, Zeye Wu, Xiuyu |
| contents | Instruct LLM provide a paradigm used in large scale language model to align LLM to human preference. The paradigm contains supervised fine tuning and reinforce learning from human feedback. This paradigm is also used in downstream scenarios to adapt LLM to specific corpora and applications. Comparing to SFT, there are many efforts focused on RLHF and several algorithms being proposed, such as PPO, DPO, IPO, KTO, MinorDPO and etc. Meanwhile most efforts for SFT are focused on how to collect, filter and mix high quality data. In this article with insight from DPO and MinorDPO, we propose a training metric for SFT to measure the discrepancy between the optimized model and the original model, and a loss function MinorSFT that can increase the training effectiveness, and reduce the discrepancy between the optimized LLM and original LLM. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_10642 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Minor SFT loss for LLM fine-tune to increase performance and reduce model deviation Xie, Shiming Chen, Hong Yu, Fred Sun, Zeye Wu, Xiuyu Artificial Intelligence Computation and Language Instruct LLM provide a paradigm used in large scale language model to align LLM to human preference. The paradigm contains supervised fine tuning and reinforce learning from human feedback. This paradigm is also used in downstream scenarios to adapt LLM to specific corpora and applications. Comparing to SFT, there are many efforts focused on RLHF and several algorithms being proposed, such as PPO, DPO, IPO, KTO, MinorDPO and etc. Meanwhile most efforts for SFT are focused on how to collect, filter and mix high quality data. In this article with insight from DPO and MinorDPO, we propose a training metric for SFT to measure the discrepancy between the optimized model and the original model, and a loss function MinorSFT that can increase the training effectiveness, and reduce the discrepancy between the optimized LLM and original LLM. |
| title | Minor SFT loss for LLM fine-tune to increase performance and reduce model deviation |
| topic | Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2408.10642 |