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.
Vivo didn't go into detail about the phone's specs, but you can likely expect a high-end processor, screen, battery and other internal components when it's fully unveiled later on. Vivo did say that like the X300 Pro, the X300 Ultra will have an impressive 200MP telephoto sensor, and is likely to offer features like 4K 120fps Dolby Vision HDR recording. Vivo's Daniel Goetz said it couldn't yet reveal the specs for its primary camera sensor — which will likely be deeply involved with the new device's push into video.
。关于这个话题，搜狗输入法2026提供了深入分析
36氪获悉，百度发布2025年第四季度及全年财报。财报显示，百度2025年总营收达1291亿元，AI业务营收达400亿元，AI云收入同比增长34%。四季度营收327亿元，上年同期营收341.24亿元；其中AI业务收入占百度一般性业务收入的43%；萝卜快跑Q4全球出行服务次数达340万，同比增长超200%；季度内每周出行次数峰值超30万。，详情可参考heLLoword翻译官方下载
:first-child]:h-full [&:first-child]:w-full [&:first-child]:mb-0 [&:first-child]:rounded-[inherit] h-full w-full
02、套壳的智能音箱？AI玩具需全新的产品思维如果只是智能对话，如今的AI玩具和智能音箱有什么区别？