Instruction Tuning with and without Context: Behavioral Shifts and Downstream Impact

TL;DR

This paper examines how instruction tuning with or without context influences large language models' behavior and downstream tasks, revealing that context affects knowledge utilization and model robustness across text and vision domains.

Contribution

It provides a comprehensive analysis of the effects of context in instruction tuning, demonstrating distinct behavioral shifts and proposing strategies for deploying more robust models.

Findings

Context-augmented training improves grounding and reduces reliance on parametric knowledge.

Using context-augmented models in vision-language tasks reduces hallucinations.

Routing inputs between separate context-aware and context-free models enhances robustness.

Abstract

Instruction tuning is a widely used approach to improve the instruction-following ability of large language models (LLMs). Instruction-tuning datasets typically include a mixture of context-augmented and context-free examples, yet prior work has largely combined these data types without examining their distinct effects. In this paper, we investigate how training LLMs with or without context affects model behavior and downstream performance. First, in the text domain, we show that LLMs trained with context attend more strongly to the provided knowledge, achieving better grounding. We also observe that context-augmented training shifts how LLMs use knowledge: models store and leverage less on parametric knowledge and instead depend more on the provided context. Second, we observe that using LLM trained with context-augmented data as the backbone for vision-language models reduces hallucination and improves grounding in the visual domain. Finally, we explore practical strategies for real-world deployments where context availability varies. We show that maintaining separate context-augmented and context-free models and routing inputs between them yields more robust overall performance than training a single mixed model, as it better preserves their complementary strengths.