Elamur Research⌕
☰
home / papers / 2025 / From RAG to Memory: Non-Parametric Continual Learning for Large Language Models

From RAG to Memory: Non-Parametric Continual Learning for Large Language Models

Paper · 2025 · Bernal Jimenez Gutierrez, Yiheng Shu, Weijian Qi et al.

Summary

Updating an LLM's knowledge through fine-tuning or model editing is expensive and prone to catastrophic forgetting. Retrieval-augmented generation (RAG) has therefore become the usual way to supply new information without changing model weights. Plain vector retrieval, however, handles two things poorly: sense-making over long, complex contexts, and associativity, meaning linking facts across several hops. Structure-augmented systems such as RAPTOR, GraphRAG, LightRAG and HippoRAG target these gaps. The authors show, though, that these systems fall behind strong embedding retrievers on simple factual QA.

HippoRAG 2 keeps HippoRAG's OpenIE-built knowledge graph and Personalized PageRank search, and changes three things:

  • it adds passage nodes directly to the graph;
  • it matches the query against whole triples rather than only extracted entities;
  • it calls an LLM at query time to discard irrelevant triples.

The authors evaluated it on factual, sense-making and multi-hop benchmarks. It beats the strongest baseline, NV-Embed-v2, by about 7 points on associative tasks and matches or slightly exceeds it elsewhere. This makes it a credible step toward non-parametric continual learning for LLMs.

Key points

  • The evaluation covers three memory types: factual (NaturalQuestions, PopQA), sense-making (NarrativeQA) and associativity (MuSiQue, 2Wiki, HotpotQA, LV-Eval).
  • In the ablations, linking the query to triples raises Recall@5 by 12.5% on average compared with the original NER-to-node linking.
  • The gains hold across dense retrievers. On a MuSiQue subset, passage recall@5 rises from 63.6 to 68.8 (GTE-Qwen2-7B-Instruct), 66.0 to 71.6 (GritLM-7B) and 69.7 to 74.7 (NV-Embed-v2).
  • In an incremental-knowledge setup (NQ, MuSiQue), the advantage over NV-Embed-v2 stays stable as data grows. Multi-hop accuracy declines for both systems, while simple QA holds steady.
  • The paper was published at ICML 2025, and code and data are available at github.com/OSU-NLP-Group/HippoRAG.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 worksInformation Systems · 1,730 works

🧭 Topics 1Research subjects and languages the work is about.Language models & pretraining · 3,894 works

Cite

Cite as Bernal Jimenez Gutierrez, Yiheng Shu, Weijian Qi, Sizhe Zhou and Yu Su. 2025. From RAG to Memory: Non-Parametric Continual Learning for Large Language Models. arXiv:2502.14802.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2502.14802 · arXiv 2502.14802

@misc{gutierrez2025rag,
  title = {{From RAG to Memory: Non-Parametric Continual Learning for Large Language Models}},
  author = {Bernal Jimenez Gutierrez, Yiheng Shu, Weijian Qi, Sizhe Zhou and Yu Su},
  year = {2025},
  doi = {10.48550/arxiv.2502.14802},
  eprint = {2502.14802},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2502.14802}
}

References

45 references, 19 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. AI@Meta. Llama 3 model card. 2024. URL https://github.com/meta-llama/llama3/blob/main/MODEL_CARD.md. ↗
  2. Beyeler, M., Rounds, E. L., Carlson, K. D., Dutt, N., and Krichmar, J. L. Neural correlates of sparse coding and dimensionality reduction. PLoS Comput Biol, 15(6): e1006908, 2019. doi: 10.1371/journal.pcbi.1006908. — [doi:10.1371/journal.pcbi.1006908](https://doi.org/10.1371/journal.pcbi.1006908) find
  3. Chen, H., Pasunuru, R., Weston, J., and Celikyilmaz, A. Walking down the memory maze: Beyond context limit through interactive reading, 2023. URL https://arxiv.org/abs/2310.05029. — [arXiv:2310.05029](https://arxiv.org/abs/2310.05029) ↗
  4. Cohen, R., Biran, E., Yoran, O., Globerson, A., and Geva, M. Evaluating the ripple effects of knowledge editing in language models. Transactions of the Association for Computational Linguistics, 12:283–298, 2024. doi: 10.1162/tacl a 00644. URL https://aclanthology.org/2024.tacl-1.16/. ↗
  5. Edge, D., Trinh, H., Cheng, N., Bradley, J., Chao, A., Mody, A., Truitt, S., and Larson, J. From local to global: A graph rag approach to query-focused summarization, 2024. URL https://arxiv.org/abs/2404.16130. ↗
  6. Gu, J.-C., Xu, H.-X., Ma, J.-Y., Lu, P., Ling, Z.-H., Chang, K.-W., and Peng, N. Model editing harms general abilities of large language models: Regularization to the rescue. In Al-Onaizan, Y., Bansal, M., and Chen, Y.-N. (eds.), Proceedings of the 2024 Conference on Empirical Methods in Natural Lan…
  7. Guo, Z., Xia, L., Yu, Y., Ao, T., and Huang, C. LightRAG: Simple and fast retrieval-augmented generation, 2024. URL https://arxiv.org/abs/2410.05779. ↗
  8. Gutiérrez, B. J., Shu, Y., Gu, Y., Yasunaga, M., and Su, Y. Hipporag: Neurobiologically inspired long-term memory for large language models. In The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024. URL https://openreview.net/forum?id=hkujvAPVsg. ↗
  9. Haveliwala, T. H. Topic-sensitive pagerank. In Lassner, D., Roure, D. D., and Iyengar, A. (eds.), Proceedings of the Eleventh International World Wide Web Conference, WWW 2002, May 7-11, 2002, Honolulu, Hawaii, USA, pp. 517–526.
  10. ACM, 2002. doi: 10.1145/511446.511513. URL https://dl.acm.org/doi/10.1145/511446.511513. ↗
  11. Hoelscher-Obermaier, J., Persson, J., Kran, E., Konstas, I., and Barez, F. Detecting edit failures in large language models: An improved specificity benchmark. In Rogers, A., Boyd-Graber, J., and Okazaki, N. (eds.), Findings of the Association for Computational Linguistics: ACL 2023, pp. 11548–11559… find
  12. Huang, J., Cui, L., Wang, A., Yang, C., Liao, X., Song, L., Yao, J., and Su, J. Mitigating catastrophic forgetting in large language models with self-synthesized rehearsal. In Ku, L.-W., Martins, A., and Srikumar, V. (eds.), Proceedings of the 62nd Annual Meeting of the Association for Computational… find
  13. Izacard, G., Caron, M., Hosseini, L., Riedel, S., Bojanowski, P., Joulin, A., and Grave, E. Unsupervised dense information retrieval with contrastive learning. Trans. Mach. Learn. Res., 2022, 2022. URL https://openreview.net/forum?id=jKN1pXi7b0. ↗
  14. Jin, X., Zhang, D., Zhu, H., Xiao, W., Li, S.-W., Wei, X., Arnold, A., and Ren, X. Lifelong pretraining: Continually adapting language models to emerging corpora. In Carpuat, M., de Marneffe, M.-C., and Meza Ruiz, I. V. (eds.), Proceedings of the 2022 Conference of the North American Chapter of the… find
  15. Khattab, O., Singhvi, A., Maheshwari, P., Zhang, Z., Santhanam, K., Vardhamanan, S., Haq, S., Sharma, A., Joshi, T. T., Moazam, H., Miller, H., Zaharia, M., and Potts, C. DSPy: Compiling declarative language model calls into self-improving pipelines. 2024.
  16. Kim, Y., Yoon, J., Ye, S., Bae, S., Ho, N., Hwang, S. J., and Yun, S.-Y. Carpe diem: On the evaluation of world knowledge in lifelong language models. In Duh, K., Gomez, H., and Bethard, S. (eds.), Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational… find
  17. Klein, G., Moon, B., and Hoffman, R. R. Making sense of sensemaking 1: Alternative perspectives. IEEE intelligent systems, 21(4):70–73, 2006. find
  18. Koli, V., Yuan, J., and Dasgupta, A. Sensemaking of socially-mediated crisis information. In Blodgett, S. L., Cercas Curry, A., Dev, S., Madaio, M., Nenkova, A., Yang, D., and Xiao, Z. (eds.), Proceedings of the Third Workshop on Bridging Human–Computer Interaction and Natural Language Processing, p… find
  19. Kwon, W., Li, Z., Zhuang, S., Sheng, Y., Zheng, L., Yu, C. H., Gonzalez, J. E., Zhang, H., and Stoica, I. Efficient memory management for large language model serving with pagedattention. In Proceedings of the ACM SIGOPS 29th Symposium on Operating Systems Principles, 2023.
  20. Lee, C., Roy, R., Xu, M., Raiman, J., Shoeybi, M., Catanzaro, B., and Ping, W. NV-embed: Improved techniques for training LLMs as generalist embedding models. In The Thirteenth International Conference on Learning Representations, 2025. URL https://openreview.net/forum?id=lgsyLSsDRe. ↗
  21. Li, J., Armandpour, M., Mirzadeh, S. I., Mehta, S., Shankar, V., Vemulapalli, R., Tuzel, O., Farajtabar, M., Pouransari, H., and Faghri, F. Tic-LM: A multi-year benchmark for continual pretraining of language models. In NeurIPS 2024 Workshop on Scalable Continual Learning for Lifelong Foundation Mod… find
  22. Li, Z., Zhang, X., Zhang, Y., Long, D., Xie, P., and Zhang, M. Towards general text embeddings with multi-stage contrastive learning. arXiv preprint arXiv:2308.03281, 2023.
  23. Liska, A., Kocisky, T., Gribovskaya, E., Terzi, T., Sezener, E., Agrawal, D., De Masson D’Autume, C., Scholtes, T., Zaheer, M., Young, S., Gilsenan-Mcmahon, E., Austin, S., Blunsom, P., and Lazaridou, A. StreamingQA: A benchmark for adaptation to new knowledge over time in question answering models.… find
  24. Lù, X. H. BM25S: Orders of magnitude faster lexical search via eager sparse scoring, 2024. URL https://arxiv.org/abs/2407.03618. — [arXiv:2407.03618](https://arxiv.org/abs/2407.03618) ↗
  25. Mallen, A., Asai, A., Zhong, V., Das, R., Khashabi, D., and Hajishirzi, H. When not to trust language models: Investigating effectiveness of parametric and non-parametric memories. In Rogers, A., Boyd-Graber, J., and Okazaki, N. (eds.), Proceedings of the 61st Annual Meeting of the Association for C… find
  26. Muennighoff, N., SU, H., Wang, L., Yang, N., Wei, F., Yu, T., Singh, A., and Kiela, D. Generative representational instruction tuning. In The Thirteenth International Conference on Learning Representations, 2025. URL https://openreview.net/forum?id=BC4lIvfSzv. ↗
  27. Ni, J., Qu, C., Lu, J., Dai, Z., Hernandez Abrego, G., Ma, J., Zhao, V., Luan, Y., Hall, K., Chang, M.-W., and Yang, Y. Large dual encoders are generalizable retrievers. In Goldberg, Y., Kozareva, Z., and Zhang, Y. (eds.), Proceedings of the 2022 Conference on Empirical Methods in Natural Language P…
  28. Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., Köpf, A., Yang, E. Z., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., Bai, J., and Chintala, S. PyTorch: An imperative style, high-pe… — PyTorch: An Imperative Style, High-Performance Deep Learning Library
  29. Robertson, S. E. and Walker, S. Some simple effective approximations to the 2-poisson model for probabilistic weighted retrieval. In Croft, W. B. and van Rijsbergen, C. J. (eds.), Proceedings of the 17th Annual International ACM-SIGIR Conference on Research and Development in Information Retrieval.… find
  30. Roth, K., Udandarao, V., Dziadzio, S., Prabhu, A., Cherti, M., Vinyals, O., Henaff, O. J., Albanie, S., Bethge, M., and Akata, Z. A practitioner’s guide to continual multimodal pretraining. In NeurIPS 2024 Workshop on Scalable Continual Learning for Lifelong Foundation Models, 2024. URL https://open… find
  31. Sarthi, P., Abdullah, S., Tuli, A., Khanna, S., Goldie, A., and Manning, C. D. RAPTOR: recursive abstractive processing for tree-organized retrieval. In The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024.
  32. OpenReview.net, 2024. URL https://openreview.net/forum?id=GN921JHCRw. ↗
  33. Shi, H., Xu, Z., Wang, H., Qin, W., Wang, W., Wang, Y., Wang, Z., Ebrahimi, S., and Wang, H. Continual learning of large language models: A comprehensive survey. arXiv preprint arXiv:2404.16789, 2024. find
  34. Suzuki, W. A. Associative learning and the hippocampus. Psychological Science Agenda, February 2005. find
  35. Thakur, N., Reimers, N., Rücklé, A., Srivastava, A., and Gurevych, I. BEIR: A heterogeneous benchmark for zero-shot evaluation of information retrieval models. In Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2), 2021. URL https://openreview.… ↗
  36. Trivedi, H., Balasubramanian, N., Khot, T., and Sabharwal, A. MuSiQue: Multihop questions via single-hop question composition. Transactions of the Association for Computational Linguistics, 10:539–554, 2022. doi: 10.1162/tacl a 00475. URL https://aclanthology.org/2022.tacl-1.31/. ↗
  37. Uner, O. and Roediger III, H. L. Do recall and recognition lead to different retrieval experiences? The American Journal of Psychology, 135(1):33–43, 2022. find
  38. Wang, Y., Ren, R., Li, J., Zhao, X., Liu, J., and Wen, J. REAR: A relevance-aware retrieval-augmented framework for open-domain question answering. In Al-Onaizan, Y., Bansal, M., and Chen, Y. (eds.), Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024,…
  39. Wolf, T., Debut, L., Sanh, V., Chaumond, J., Delangue, C., Moi, A., Cistac, P., Rault, T., Louf, R., Funtowicz, M., and Brew, J. Huggingface’s transformers: State-of-the-art natural language processing. CoRR, abs/1910.03771, 2019. URL http://arxiv.org/abs/1910.03771. ↗
  40. Xie, J., Zhang, K., Chen, J., Lou, R., and Su, Y. Adaptive chameleon or stubborn sloth: Revealing the behavior of large language models in knowledge conflicts. In The Twelfth International Conference on Learning Representations, 2024. URL https://openreview.net/forum?id=auKAUJZMO6. ↗
  41. Yao, Y., Wang, P., Tian, B., Cheng, S., Li, Z., Deng, S., Chen, H., and Zhang, N. Editing large language models: Problems, methods, and opportunities. In Bouamor, H., Pino, J., and Bali, K. (eds.), Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pp. 10222–1024… find
  42. Yuan, T., Ning, X., Zhou, D., Yang, Z., Li, S., Zhuang, M., Tan, Z., Yao, Z., Lin, D., Li, B., Dai, G., Yan, S., and Wang, Y. LV-Eval: A balanced long-context benchmark with 5 length levels up to 256k, 2024. URL https://arxiv.org/abs/2402.05136. ↗
  43. Zhang, H., Gui, L., Zhai, Y., Wang, H., Lei, Y., and Xu, R. Copr: Continual learning human preference through optimal policy regularization, 2024. URL https://arxiv.org/abs/2310.15694. ↗
  44. Zhang, Z., Fang, M., Chen, L., and Namazi-Rad, M.-R. CITB: A benchmark for continual instruction tuning. In Bouamor, H., Pino, J., and Bali, K. (eds.), Findings of the Association for Computational Linguistics: EMNLP 2023, pp. 9443–9455, Singapore, December 2023. Association for Computational Lingui… find
  45. Zhong, Z., Wu, Z., Manning, C., Potts, C., and Chen, D. MQuAKE: Assessing knowledge editing in language models via multi-hop questions. In Bouamor, H., Pino, J., and Bali, K. (eds.), Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pp. 15686–15702, Singapore, D… find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/gutierrez-2025-rag-memory-non-parametric-continual/