The story’s score and claims

Comments on No. 33 of 143 · arXiv · edition of Wednesday 11 November 2026

MemoryDojo: Benchmarking Poisoning Through LLM Agent Memory

Summary

MemoryDojo is a benchmark for measuring how poisoned instructions persist and spread through LLM agent memory systems across multi-session tasks. The authors' 155 trajectories across four domains and five memory systems show all tested architectures are vulnerable, with notable differences in downstream effects.

Comments

Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules