ImpRIF: Stronger Implicit Reasoning Leads to Better Complex Instruction Following

TL;DR

This paper introduces ImpRIF, a method that improves large language models' ability to follow complex instructions involving implicit reasoning by formalizing instructions as verifiable reasoning graphs and training models accordingly.

Contribution

ImpRIF formalizes implicit reasoning instructions as verifiable graphs and employs graph-driven training to enhance LLMs' complex instruction following capabilities.

Findings

Models with ImpRIF outperform base models on five benchmarks.

Synthesized large-scale data improves reasoning accuracy.

Reinforcement learning further enhances implicit reasoning skills.

Abstract

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.