Abstract:Humans shift between different personas depending on social context. Large Language Models (LLMs) demonstrate a similar flexibility in adopting different personas and behaviors. Existing approaches, however, typically adapt such behavior through external knowledge such as prompting, retrieval-augmented generation (RAG), or fine-tuning. We ask: do LLMs really need external context or parameters to adapt to different behaviors, or do they already have such knowledge embedded in their parameters? In this work, we show that LLMs already contain persona-specialized subnetworks in their parameter space. Using small calibration datasets, we identify distinct activation signatures associated with different personas. Guided by these statistics, we develop a masking strategy that isolates lightweight persona subnetworks. Building on the findings, we further discuss: how can we discover opposing subnetwork from the model that lead to binary-opposing personas, such as introvert-extrovert? To further enhance separation in binary opposition scenarios, we introduce a contrastive pruning strategy that identifies parameters responsible for the statistical divergence between opposing personas. Our method is entirely training-free and relies solely on the language model's existing parameter space. Across diverse evaluation settings, the resulting subnetworks exhibit significantly stronger persona alignment than baselines that require external knowledge while being more efficient. Our findings suggest that diverse human-like behaviors are not merely induced in LLMs, but are already embedded in their parameter space, pointing toward a new perspective on controllable and interpretable personalization in large language models.
In 1950, Bank of America approached SRI about the feasibility of an automated
，详情可参考im钱包官方下载
Then, over the last week, online researchers found that Persona, another company Discord had partnered with in the UK for age verification, had left thousands of files exposed on the open internet.
Moment of introspection aside, I’m not sure what the future holds for agents and generative AI. My use of agents has proven to have significant utility (for myself at the least) and I have more-than-enough high-impact projects in the pipeline to occupy me for a few months. Although certainly I will use LLMs more for coding apps which benefit from this optimization, that doesn’t imply I will use LLMs more elsewhere: I still don’t use LLMs for writing — in fact I have intentionally made my writing voice more sardonic to specifically fend off AI accusations.
Сайт Роскомнадзора атаковали18:00