Long-Horizon Memory

Cite this benchmark

@misc{zhao2026lmeb,
  archiveprefix = {arXiv},
  author = {Zhao, Xinping and Hu, Xinshuo and Xu, Jiaxin and Tang, Danyu and Zhang, Xin and Zhou, Mengjia and Zhong, Yan and Zhou, Yao and Shan, Zifei and Zhang, Meishan and Hu, Baotian and Zhang, Min},
  eprint = {2603.12572},
  primaryclass = {cs.CL},
  title = {LMEB: Long-horizon Memory Embedding Benchmark},
  url = {https://arxiv.org/abs/2603.12572},
  year = {2026},
}

Long-horizon memory retrieval quality across episodic, dialogue, semantic, and procedural retrieval tasks, measuring how well embedding models retrieve evidence in long-term memory scenarios.

Reference paper →

Languages 1

Tasks 22

Task Types 1

Models 0