Designing Role Vectors to Improve LLM Inference Behaviour
This work addresses improving LLM inference behavior for AI practitioners, but it is incremental as it builds on existing persona studies.
The paper tackled the problem of uncertain impact of personas on LLM performance by exploring role vectors as an alternative, finding they improved task performance in relevant domains with minimal effect on unrelated tasks.
The influence of personas on Large Language Models (LLMs) has been widely studied, yet their direct impact on performance remains uncertain. This work explores a novel approach to guiding LLM behaviour through role vectors, an alternative to persona-based prompting. We construct 29 role vectors derived from model activations and evaluate their impact on benchmark performance across multiple domains. Our analysis investigates whether these vectors can effectively steer models toward domain-specific expertise. We measure two key interventions: (i) activation addition, which reinforces role-specific directions, and (ii) directional ablation, which removes them. Results on well-established benchmarks indicate that role vectors do, in fact, influence model behaviour, improving task performance in relevant domains while marginally affecting unrelated tasks. This, in turn, suggests that manipulating internal model representations has a greater impact on outcomes than persona-based prompting.