他们搞了个带记忆的 8B 多模态智能体，20 轮对话检索准确率 91.4%，比 32B 模型还快一半。代码和 demo 都开源了，值得看。
该论文提出 Cognitive-structured Multimodal Agent (CMA)，通过外化视觉信息至 Episodic Visual Memory 并在推理时选择性激活相关片段，解决统一多模态模型在长时对话中视觉 token 爆炸和跨轮引用不可靠的问题。CMA 由 Perceptual Abstraction Engine、Cognitive Retrieval Engine 和 Multimodal Executive Controller 组成。利用 Unified Scenario Engine 生成带细粒度检索标注的结构化多轮对话，并通过强化学习优化策略。在自主构建的长程视觉对话基准上，8B 参数的 CMA 在 20 轮会话中达到 91.4% 检索准确率，比 32B 基线高 8.2%，每轮推理时间从 23.1 秒降至 12.7 秒。
Cognitive-structured Multimodal Agent for Multimodal Understanding, Generation, and Editing
Recent unified multimodal models show a single architecture can jointly perform vision/language understanding and image generation/editing. However, they repeatedly feed all historical visual and textual inputs into a shared context window, limiting long-horizon multimodal dialogue due to visual token explosion and unreliable cross-turn referencing. We propose a Cognitive-structured Multimodal Agent that externalizes visual information into an Episodic Visual Memory and selectively reactivates relevant episodes during reasoning. The agent consists of a Perceptual Abstraction Engine for structured visual abstraction, a Cognitive Retrieval Engine for cross-turn memory retrieval, and a Multimodal Executive Controller for autonomous task inference and action planning. To address the lack of turn-level retrieval supervision in existing datasets, we develop a Unified Scenario Engine that programmatically generates structured multi-turn conversations with fine-grained retrieval annotations, enabling reinforcement learning to optimize abstraction and retrieval policies. We also construct a long-horizon visual-dialogue benchmark stratified by difficulty to evaluate episodic visual recall. Our 8B agent achieves 91.4% retrieval accuracy over 20-turn sessions, surpassing 32B baselines by +8.2% while nearly halving per-turn inference time (23.1s -> 12.7s). We further present the Cognitive-structured Multimodal Agent Harness (CMA-Harness), a tool-augmented deployment of the same cognitive structure integrating persistent multimodal memory, web access, image generation/editing/composition tools, and OpenAI-compatible serving. Structured memory and modular decision-making offer a more scalable, efficient paradigm for long-horizon multimodal agents than monolithic parameter scaling. Code: https://github.com/caseclose/cma-harness ; Project page: https://caseclose.github.io/cma-harness/