MIMESIS is a purpose-built user simulator trained on human conversations with explicit reasoning supervision and 13 realistic behavioral patterns derived from real user interactions. Given a user’s goal, it plays the user in a conversation with an AI assistant and thinks privately before each message. MIMESIS-4B is fine-tuned from Qwen3.5-4B. A 9B version is also available. Scores for MIMESIS and frontier API models. Lower is better for Turing distance. The model was trained with the roles swapped: the assistant under test speaks in the user role, and MIMESIS replies in the assistant role. Its chat template opens a block, so each completion holds a private thought, then , then the message. Loading requires transformers 5.2 or later. To continue, append the full completion as an assistant message and the assistant’s reply as a user message. Allow a generous token budget, since the thought comes first. With vLLM or SGLang, keep the model’s own chat template. MIMESIS is a research model for training and evaluating interactive agents. It imitates user behavior, including impatience, vagueness, and non-cooperation, so it is not an assistant.
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: phanviethoang1512
Теги: qwen3_5, image-text-to-text, user-simulation, conversational, endpoints_compatible
Лайков: 3 | Загрузок: 844
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
