cs.CLNov 16, 2026

MemDoctor: Stage-wise Error Attribution and Repair for MLLM Agent Memory

Authors: Ruiyang Jiang, Xinyang Liao, Weikai Huang, Yuqi Niu, Junjie Zhou, Yubin Zheng

Organizations: Shanghai Jiao Tong University · Xi’an Jiaotong University · Xidian University

Abstract

Multimodal large language models (MLLMs) now assist users across sessions of images and text, carried by an external memory of what was said and shown. When such an assistant answers wrongly, the failure usually began earlier, in how a record was written, updated or retrieved, and the answer alone does not say where. We formalize multimodal memory as four stages, Write, Update, Retrieve and Answer, and define nine error types across them. A stage-resolved analysis on Mem-Gallery traces most failures to a source stage, and most of those lie before the answer. Injecting known errors against fact-preserving control edits, we then follow each tainted record downstream. Early errors survive to the answer as apparent evidence, and hurt most when no other record holds the fact. A diagnosis that trusts the quoted evidence therefore blames the stage where the error surfaces, not the one where it began. We propose MemDoctor, in which the MLLM that built the memory diagnoses its own failures. It traces each needed fact backward through the four stages, accepts a diagnosis only when the cited evidence is found in memory, and repairs that stage before answering again. On a 4B backbone, MemDoctor improves accuracy on both Mem-Gallery and MemLens, outperforming all evaluated 8B baselines, and the same loop also improves MuRAG and A-Mem.

Figures & tables

Appendix figures & tables7 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. EvalMem: An Operation-Level Diagnostic Framework for Long-Term Memory Systems

    Sep 3, 2026Zeyu Liu, Jian Zhong, Rongduo Han +12LLM MemoryLLM Agent Memory

  2. When Errors Become Memories: Causal Pathway Tracing in Multi-Turn Memory-Augmented LLMs

    Aug 31, 2026Shuyao Xiao, Shengling Wang, Xuan Chen +8error propagation in large language model inferenceMemory-Augmented Language Models

  3. MemFail: Stress-Testing Failure Modes of LLM Memory Systems

    May 26, 2026Ishir Garg, Neel Kolhe, Dawn Song +1LLM Agent MemoryLLM Memory