FadeMem: Biologically-Inspired Forgetting for Efficient Agent Memory
Abstract
Large language models deployed as autonomous agents face critical memory limitations, lacking selective forgetting mechanisms that lead to either catastrophic forgetting at context boundaries or information overload within them. While human memory naturally balances retention and forgetting through adaptive decay processes, current AI systems employ binary retention strategies that preserve everything or lose it entirely. We propose FadeMem, a biologically-inspired agent memory architecture that incorporates active forgetting mechanisms mirroring human cognitive efficiency. FadeMem implements differential decay rates across a dual-layer memory hierarchy, where retention is governed by adaptive exponential decay functions modulated by semantic relevance, access frequency, and temporal patterns. Through LLM-guided conflict resolution and intelligent memory fusion, our system consolidates related information while allowing irrelevant details to fade. Experiments on Multi-Session Chat, LoCoMo, and LTI-Bench demonstrate superior multi-hop reasoning and retrieval with 45% storage reduction, validating the effectiveness of biologically-inspired forgetting in agent memory systems.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Effective memory requires selective retention and forgetting, as retaining all information leads to performance degradation.
“Prior work in continual learning and neural memory systems has shown that effective memory requires mechanisms for selective retention and forgetting, as retaining all information leads to performance degradation Kirkpatrick et al. [2017], Wei et al. [2026].”
From Selective Forgetting: A Graph-Based Memory Framework for Long-Term LLM Agents (2026) · cites this paper as Wei et al. [2026]
FadeMem introduces biologically inspired forgetting to keep agent memory efficient.
“FadeMem Wei et al. [2026] introduces biologically inspired forgetting to keep agent memory efficient.”
From Selective Forgetting: A Graph-Based Memory Framework for Long-Term LLM Agents (2026) · cites this paper as Wei et al. [2026]
Cites
Overcoming catastrophic forgetting in neural networks
Kirkpatrick, J., Pascanu, R., Rabinowitz, N., Veness, J., Desjardins, G., Rusu, A. A., Milan, K., Quan, J., Ramalho, T., Grabska-Barwinska, A., et al. (2017) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1611835114 · Cited by 7,899 · see citing works
Dense passage retrieval for open-domain question answering
Karpukhin, V., Oguz, B., Min, S., Lewis, P., Wu, L., Edunov, S., Chen, D., & Yih, W.-T. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2020.emnlp-main.550 · Cited by 3,575 · see citing works
Transformer-XL: Attentive language models beyond a fixed-length context
Dai, Z., Yang, Z., Yang, Y., Carbonell, J., Le, Q., & Salakhutdinov, R. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1285 · Cited by 3,169 · see citing works
Lost in the Middle: How Language Models Use Long Contexts
Liu, N. F., Lin, K., Hewitt, J., Paranjape, A., Bevilacqua, M., Petroni, F., & Liang, P. (2024) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00638 · Cited by 1,612 · see citing works
The Psychology and Neuroscience of Forgetting
Wixted, J. T. (2004) · Annual Review of Psychology · doi:10.1146/annurev.psych.55.090902.141555 · Cited by 999 · see citing works
SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models
Manakul, P., Liusie, A., & Gales, M. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.557 · Cited by 557 · see citing works
Beyond Goldfish Memory: Long-Term Open-Domain Conversation
Xu, J., Szlam, A., & Weston, J. (2022) · Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2022.acl-long.356 · Cited by 188 · see citing works
Beyond Goldfish Memory: Long-Term Open-Domain Conversation
Xu, J., Szlam, A., & Weston, J. (2021) · arXiv · doi:10.48550/arxiv.2107.07567 · Cited by 148 · see citing works
A Survey on the Memory Mechanism of Large Language Model-based Agents
Zhang, Z., Dai, Q., Bo, X., Ma, C., Li, R., Chen, X., Zhu, J., Dong, Z., & Wen, J.-R. (2025) · ACM Transactions on Information Systems · doi:10.1145/3748302 · Cited by 118 · see citing works
Evaluating Very Long-Term Conversational Memory of LLM Agents
Maharana, A., Lee, D.-H., Tulyakov, S., Bansal, M., Barbieri, F., & Fang, Y. (2024) · Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2024.acl-long.747 · Cited by 96 · see citing works
Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
Chhikara, P., Khant, D., Aryan, S., Singh, T., & Yadav, D. (2025) · Frontiers in Artificial Intelligence and Applications · doi:10.3233/faia251160 · Cited by 69 · see citing works
Catastrophic Forgetting in Deep Learning: A Comprehensive Taxonomy
Aleixo, E. L., Colonna, J. G., Cristo, M., & Fernandes, E. (2024) · Journal of the Brazilian Computer Society · doi:10.5753/jbcs.2024.3966 · Cited by 54 · see citing works
Catastrophic Forgetting in Deep Learning: A Comprehensive Taxonomy
Aleixo, E. L., Colonna, J. G., Cristo, M., & Fernandes, E. (2023) · arXiv · doi:10.48550/arxiv.2312.10549 · Cited by 54
Kang, J., Ji, M., Zhao, Z., & Bai, T. (2025) · Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2025.emnlp-main.1318 · Cited by 23 · see citing works
Akbar, N. A., Dembani, R., Lenzitti, B., & Tegolo, D. (2025) · IEEE Access · doi:10.1109/access.2025.3589241 · Cited by 18 · see citing works
Machine Memory Intelligence: Inspired by Human Memory Mechanisms
Zheng, Q., Liu, H., Zhang, X., Yan, C., Cao, X., Gong, T., Liu, Y., Shi, B., Peng, Z., Fan, X., et al. (2025) · Engineering · doi:10.1016/j.eng.2025.01.012 · Cited by 14 · see citing works
Multimodal Multi-Agent Empowered Legal Judgment Prediction
Kang, Z., Gong, J., Qingxi, C. H. E. N., Zhang, H., Liu, J., Fu, R., Feng, Z., Wang, Y., Fong, S., & Zhou, K. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11462917 · Cited by 12 · see citing works
Rudroff, T., Rainio, O., & Klén, R. (2024) · Brain Sciences · doi:10.3390/brainsci14111111 · Cited by 11 · see citing works
Bio-inspired cognitive architecture of episodic memory
Martin, L., Jaime, K., Ramos, F., & Robles, F. (2022) · Cognitive Systems Research · doi:10.1016/j.cogsys.2022.08.001 · Cited by 9 · see citing works
Biologically Inspired Sleep Algorithm for Reducing Catastrophic Forgetting in Neural Networks
Tadros, T., Krishnan, G., Ramyaa, R., & Bazhenov, M. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i10.7239 · Cited by 9 · see citing works
Large Language Model Agents: A Comprehensive Survey on Architectures, Capabilities, and ApplicationsPreprint
Lei, Y., Xu, J., Liang, C. X., Bi, Z., Li, X., Zhang, D., Song, J., & Yu, Z. (2025) · Preprints.org · doi:10.20944/preprints202512.2119.v1 · Cited by 4 · see citing works
A biologically inspired architecture with switching units can learn to generalize across backgrounds
Voina, D., Shea-Brown, E., & Mihalas, S. (2023) · Neural Networks · doi:10.1016/j.neunet.2023.09.014 · Cited by 2 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Multimodal Multi-Agent Empowered Legal Judgment Prediction
Kang, Z., Gong, J., Qingxi, C. H. E. N., Zhang, H., Liu, J., Fu, R., Feng, Z., Wang, Y., Fong, S., & Zhou, K. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11462917 · Cited by 12 · see citing works
Sovereign Epistemic Commons under No-Meta GovernancePreprint
Takahashi, K. (2026) · Zenodo (CERN European Organization for Nuclear Research) · doi:10.5281/zenodo.18997827 · Cited by 2 · see citing works
Selective Forgetting: A Graph-Based Memory Framework for Long-Term LLM Agents
Rusu, T., Khanzadeh, S., & Alalfi, M. (2026) · arXiv · doi:10.48550/arxiv.2608.28978
PyRAMIND: Auditable Deterministic Pyramid Memory for Long-Running LLM Agents
Roy, A., Upadhyay, R., Millar, A. P., Murray, L., & Mills, R. (2026) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3799682.3840047
H-Mem: Hawkes-driven Adaptive Forgetting for Long-Horizon Agent Memory
Wu, X., Qin, H., Zeng, G., Li, R.-H., & Zhao, K. (2026) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3799682.3840962
- Authors
- Wei, L., Dong, X., Peng, X., Xie, N., & Wang, B.
- Affiliations
- Alibaba Group (China) (CN)
- Peking University (CN)
- ORCID
- Wang, Bin
- Venue
- ICASSP 2026 - 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
- Publisher
- IEEE
- Published
- 2026-05-03
- Pages
- 4011-4015
- ISSN
- 1520-6149 (ISSN-L)
- OpenAlex
- W7155058168
- Cited by
- 1
- References
- 20
- Citations by year
- 2026: 1
- FWCI
- 8.94
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE International Conference on Acoustics Speech and Signal Processing (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › AI-based Problem Solving and Planning
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
- Keywords
- agent memory, catastrophic forgetting, memory hierarchy, large language models, autonomous agents, memory consolidation, multi-hop reasoning, retrieval, semantic relevance, access frequency, temporal patterns
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
