Training with Pseudo-Code for Instruction Following
This addresses the issue of unreliable instruction-following in LLMs for users, but it is incremental as it builds on existing pseudo-code methods.
The paper tackled the problem of large language models struggling to follow simple instructions, especially compositions, by fine-tuning them with instruction-tuning data that includes instructions re-expressed in pseudo-code, resulting in relative gains of 3–19% on instruction-following benchmarks and up to 14% average gains across tasks.
Despite the rapid progress in the capabilities of Large Language Models (LLMs), they continue to have difficulty following relatively simple, unambiguous instructions, especially when compositions are involved. In this paper, we take inspiration from recent work that suggests that models may follow instructions better when they are expressed in pseudo-code. However, writing pseudo-code programs can be tedious and using few-shot demonstrations to craft code representations for use in inference can be unnatural for non-expert users of LLMs. To overcome these limitations, we propose fine-tuning LLMs with instruction-tuning data that additionally includes instructions re-expressed in pseudo-code along with the final response. We evaluate models trained using our method on $11$ publicly available benchmarks comprising of tasks related to instruction-following, mathematics, and common-sense reasoning. We conduct rigorous experiments with $5$ different models and find that not only do models follow instructions better when trained with pseudo-code, they also retain their capabilities on the other tasks related to mathematical and common sense reasoning. Specifically, we observe a relative gain of $3$--$19$% on instruction-following benchmark, and an average gain of upto 14% across all tasks.