Conference paper · 2026 · Open access
SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents
Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP), 25309–25359 · Published 1 January 2026 · DOI 10.18653/2026.findings-emnlp.1199
In brief
“SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents” is a 2026 conference paper by Wenxuan Wang and 6 co-authors, published in Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP). OpenAlex counts 2 citations as of 19 November 2026, which puts it in the top 6% of works published in 2026 by citations. It is open access: the full text is free to read at Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP).
- Citations
- 2
- Most cited in
- 2026 (2)
- Field-weighted impact
- 1.77
- References
- 36
- Authors
- 7
Abstract
Persistent AI assistants, such as OpenClaw, accumulate large collections of related memories over long-term interactions.As these memories grow, they may reinforce one another, diverge across contexts, or directly conflict, making correct assistance depend on memory relations rather than isolated recall.Existing long-term memory benchmarks do not systematically probe how agents preserve and utilize such relations during downstream tasks.To address this gap, we introduce SubtleMemory, a benchmark for fine-grained relational memory discrimination in long-running AI agents.SubtleMemory constructs relation-controlled latent semantic artifacts whose variants instantiate complementary, nuanced, or contradictory relations, and embeds them into realistic user-agent histories, requiring agents to recover distributed relational structures during later queries and instructions.The benchmark contains 1,522 evaluation instances over 10 long histories, grounded in 1,090 relationcontrolled memory-variant sets and spanning user-related and non-user-related queries.Evaluating six standalone memory systems, two Claw-style agents with native memory modules, and three Claw-style agents with plugin memory modules, we find that current systems remain weak on fine-grained relational memory discrimination.We further introduce diagnostic protocols that reveal distinct capability profiles across memory preservation, retrieval, and downstream reasoning stages.Code and data are publicly available.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Wang, W., Sun, H., Hou, F., Song, M., Zhang, W., Cheng, Y., & (45629), Y. Y. (2026). SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents. Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP), 25309–25359. https://doi.org/10.18653/2026.findings-emnlp.1199
MLA 9
Wang, Wenxuan, et al. “SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents.” Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP), 2026, pp. 25309–25359. https://doi.org/10.18653/2026.findings-emnlp.1199.
Chicago
Wang, Wenxuan, Haoyu Sun, Fukuan Hou, Mingyang Song, Weinan Zhang, Yu Cheng, and Yang Yang (45629). 2026. “SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents.” Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP): 25309–25359. https://doi.org/10.18653/2026.findings-emnlp.1199.
Harvard
Wang, W. et al. (2026) ‘SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents’, Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP), pp. 25309–25359. doi:10.18653/2026.findings-emnlp.1199.
IEEE
W. Wang et al., “SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents,” Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP), pp. 25309–25359, 2026, doi: 10.18653/2026.findings-emnlp.1199.
BibTeX
@inproceedings{wang2026subtlememory,
title = {{SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents}},
author = {Wang, Wenxuan and Sun, Haoyu and Hou, Fukuan and Song, Mingyang and Zhang, Weinan and Cheng, Yu and (45629), Yang Yang},
booktitle = {Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)},
year = {2026},
pages = {25309--25359},
doi = {10.18653/2026.findings-emnlp.1199},
url = {https://doi.org/10.18653/2026.findings-emnlp.1199}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Computer Science › Artificial Intelligence › Speech and dialogue systems
- long-term memory
- relational memory
- memory discrimination
- AI agents
- memory conflicts
- memory retrieval
- memory preservation
- diagnostic protocol
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Wang, W., Sun, H., Hou, F., Song, M., Zhang, W., Cheng, Y., & (45629), Y. Y. (2026). SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents. Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP), 25309–25359. https://doi.org/10.18653/2026.findings-emnlp.1199
How many times has this paper been cited?
2 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP): sign in and open the PDF from this page.
Who wrote this paper?
Wenxuan Wang, Haoyu Sun, Fukuan Hou, Mingyang Song, Weinan Zhang, Yu Cheng, Yang Yang (45629).
What is the DOI of this paper?
10.18653/2026.findings-emnlp.1199, permanently at https://doi.org/10.18653/2026.findings-emnlp.1199. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.