Paper Detail

ThinkFlow: Self-Evolving Probabilistic Latent Memory for Lifelong Conversational Agents

Cai Ke, Xin Liu, Han Zhang, Jiangyue Yan, Zike Yuan, Ling Deng, Yue Yu, Hui Wang, Ruifeng Xu

arxiv Score 13.8

Published 2026-09-15 · First seen 2026-09-17

General AI

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 \textbf{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.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
now
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{ke2026thinkflow,
  title = {ThinkFlow: Self-Evolving Probabilistic Latent Memory for Lifelong Conversational Agents},
  author = {Cai Ke and Xin Liu and Han Zhang and Jiangyue Yan and Zike Yuan and Ling Deng and Yue Yu and Hui Wang and Ruifeng Xu},
  year = {2026},
  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 la},
  url = {https://arxiv.org/abs/2609.17010},
  keywords = {cs.AI, cs.CL},
  eprint = {2609.17010},
  archiveprefix = {arXiv},
}

Metadata

{}