ThinkFlow: Self-Evolving Probabilistic Latent Memory for Lifelong Conversational Agents
cs.AI, cs.CL
Submitted: 2026-09-15
Updated: 2026-09-15
License: http://creativecommons.org/licenses/by/4.0/
The gist: Lifelong conversational agents rely on memory systems to maintain deep, context-aware interactions with users.
Terminology
Abstract
Lifelong conversational agents rely on memory systems to maintain deep, context-aware interactions with users. However, existing explicit textual memory pipelines suffer from a severe information bottleneck, often losing subtle behavioral patterns and emotional shifts. Furthermore, being typically static post-deployment, they cannot autonomously adapt to personal habits and preferences without manual feedback. Cognitive science, however, suggests that humans maintain mental models purely in a latent space and continuously refine them through predictive coding. Inspired by this, we propose ThinkFlow, a novel end-to-end latent memory framework for lifelong conversational agents. ThinkFlow bypasses the text bottleneck by dynamically compressing conversational flows into probabilistic latent memory skills, autonomously consolidating complex user states into disentangled, continuous vectors without semantic interference. To break this barrier, we introduce a test-time evolution paradigm. By coupling teacher-guided latent alignment to bootstrap the initial state with a self-supervised next-user-utterance prediction task for continuous refinement, the framework successfully overcomes cold-start challenges and achieves label-free lifelong personalization. Extensive experiments on long-term conversation benchmarks demonstrate that ThinkFlow significantly outperforms prevailing memory systems, providing highly personalized and contextually accurate responses over extended multi-session interactions.
Sources
- Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
- From Local to Global: A Graph RAG Approach to Query-Focused Summarization
- The Llama 3 Herd of Models
- Memory in the Age of AI Agents
- PersonaMem-v2: Towards Personalized Intelligence via Learning Implicit User Personas and Agentic Memory
- AgentLite: A Lightweight Library for Building and Advancing Task-Oriented LLM Agent System
- MemoChat: Tuning LLMs to Use Memos for Consistent Long-Range Open-Domain Conversation
- AIOS: LLM Agent Operating System
- Predictive Coding: a Theoretical and Experimental Review
- Seek in the Dark: Reasoning via Test-Time Instance-Level Policy Gradient in Latent Space
- MemGPT: Towards LLMs as Operating Systems
- From Single to Multi-Granularity: Toward Long-Term Memory Association and Selection of Conversational Agents
- Qwen3 Technical Report
- The Latent Space: Foundation, Evolution, Mechanism, Ability, and Outlook
- BERTScore: Evaluating Text Generation with BERT
- NextMem: Towards Latent Factual Memory for LLM-based Agents
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection