Tuna: Instruction Tuning using Feedback from Large Language Models
This work addresses the challenge of enhancing instruction-tuned LLMs for better alignment with human preferences, offering a cost-effective method that is incremental but shows strong gains on specific tasks.
The paper tackles the problem of instruction-tuned LLMs lacking knowledge of potentially better responses by proposing Tuna, a model finetuned using probabilistic and contextual ranking approaches, which consistently improves performance on benchmarks like Super Natural Instructions and LMentry, even outperforming several reinforcement learning baselines.
Instruction tuning of open-source large language models (LLMs) like LLaMA, using direct outputs from more powerful LLMs such as Instruct-GPT and GPT-4, has proven to be a cost-effective way to align model behaviors with human preferences. However, the instruction-tuned model has only seen one response per instruction, lacking the knowledge of potentially better responses. In this paper, we propose finetuning an instruction-tuned LLM using our novel \textit{probabilistic ranking} and \textit{contextual ranking} approaches to increase the likelihood of generating better responses. Probabilistic ranking enables the instruction-tuned model to inherit the relative rankings of high-quality and low-quality responses from the teacher LLM. On the other hand, learning with contextual ranking allows the model to refine its own response distribution using the contextual understanding ability of stronger LLMs. Furthermore, we apply probabilistic ranking and contextual ranking sequentially to the instruction-tuned LLM. The resulting model, which we call \textbf{Tuna}, consistently improves the performance on Super Natural Instructions (119 test tasks), LMentry (25 test tasks), Vicuna QA, and can even obtain better results than several strong reinforcement learning baselines. Our code and data are available at \url{ https://github.com/microsoft/LMOps}.