Multimodal Agent MemorySubmit a paper

Agentic ASR

Towards Human-Like Interactive Speech Recognition With Agentic Correction and Semantic Evaluation

Zixuan Jiang, Yanqiao Zhu, Peng Wang, Qinyuan Chen, Xinjian Zhao, Xipeng Qiu, et al. · 2026-05-28

Read the paperPDF

Management
T2 Dynamic Memory
Memory changes during deployment through a stable management policy.
Primary modality
Audio & speech
Also: text
Representation
Hybrid
Distinct operative representation paths are deliberately linked or jointly consumed by downstream reasoning, generation, or control.

How experience becomes memory

Each path is one memory component, traced from what is stored to how it is found again.

  1. updated textual transcription state and correction history

    1. Represented asPerception-to-text records & summariesCross-represented explicit
    2. Stored inFlat sequential records
    3. Retrieved byPlanner / tool-assisted retrieval
    4. Source evidenceSource not retained
    Stateful transcription history and LLM-applied edits. reasoning module uses interaction history to infer the intended edit.Full paper: B. Agentic ASR Framework · PDF p. 3 · Figure 1 / method overview

  2. updated textual transcription state and correction history

    1. Represented asPersistent multimodal latent stateInternalized latent & parametric
    2. Stored inFlat sequential records
    3. Retrieved byPlanner / tool-assisted retrieval, Latent / parametric activation
    4. Source evidenceSource not retained
    Stateful transcription history and LLM-applied edits. reasoning module uses interaction history to infer the intended edit.Full paper: B. Agentic ASR Framework · PDF p. 3 · Figure 1 / method overview

How memory changes (T2 operations)

  • Update · Prompted LLM/VLM record update

    During operation, Agentic ASR applies LLM-mediated memory editing to updated textual transcription state and correction history; the memory state changes while the governing update policy remains fixed.

Related systems