ImpRIF: Stronger Implicit Reasoning Leads to Better Complex Instruction Following
For LLM developers, this work shows that improving implicit reasoning can significantly boost complex instruction following, offering a practical approach to a known bottleneck.
The paper addresses the challenge of complex instruction following in LLMs by enhancing implicit reasoning. The proposed ImpRIF method, which uses reasoning graphs and graph-driven chain-of-thought, achieves substantial improvements over base models on five benchmarks.
As applications of large language models (LLMs) become increasingly complex, the demand for robust complex instruction following capabilities is growing accordingly. We argue that a thorough understanding of the instruction itself, especially the latent reasoning structure embedded between the lines, is crucial for improving instruction following. Therefore we target complex instructions that involve implicit reasoning, intricate logical relations, and multi-constraint dependencies. We propose ImpRIF, a method to enhance LLMs' understanding of implicit reasoning instructions, thereby improving its ability to follow complex instructions. We formalize such instructions as verifiable reasoning graphs, enabling programmatic verification and graph-driven chain-of-thought reasoning. Based on this formulation, we synthesize large-scale single- and multi-turn data, propose fine-tuning with graph reasoning, and apply reinforcement learning to explicitly train models to reason along the graph. On five complex instruction following benchmarks, our models substantially outperform their base models. These results demonstrate that enhancing implicit reasoning capabilities can significantly improve complex instruction following.