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.
The SETBLOCK function is somewhat like realloc() in the Standard C library, but never moves the allocated block. Resizing a block to the same or smaller size will always succeed, and will free up the remaining memory. Resizing to a larger size may fail, and if it does, the maximum available size will be returned in the BX register (just like when allocating).
OpenRouter的后台数据也佐证了这种疯狂，最新数据显示，OpenClaw已是OpenRouter最大的单一应用，其Token消耗量约占平台显著比例。Token消耗从"按次"变成了"按流量"，AI使用的成本曲线正在急剧变陡。
这些数据不仅是技术迭代的核心资产，更因其高度涉及其运行环境的底层信息，极易触及国家安全与公共利益的敏感神经。
In the previous example we considered a space which has only one road looping back on itself. The number of times you would walk around the road to get back to the “same” point (or an equivalent point in a different copy) can be encoded using this “winding number” trick: