∞-former: Infinite Memory Transformer-former: Infinite Memory Transformer
Open accessAbstract
Transformers are unable to model long-term memories effectively, since the amount of computation they need to perform grows with the context length. While variations of efficient transformers have been proposed, they all have a finite memory capacity and are forced to drop old information. In this paper, we propose the ∞-former, which extends the vanilla transformer with an unbounded long-term memory. By making use of a continuous-space attention mechanism to attend over the long-term memory, the ∞-former's attention complexity becomes independent of the context length, trading off memory length with precision.In order to control where precision is more important, ∞-former maintains "sticky memories," being able to model arbitrarily long contexts while keeping the computation budget fixed.Experiments on a synthetic sorting task, language modeling, and document grounded dialogue generation demonstrate the ∞-former's ability to retain information from long sequences.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
In conversational AI, well-designed memory structures are essential for maintaining personalized interactions across multiple sessions.
“In conversational AI, well-designed memory structures are essential for maintaining personalized interactions across multiple sessions (Martins et al., 2022; Bae et al., 2022; Gutiérrez et al., 2024).”
From Pre-Storage Reasoning for Episodic Memory: Shifting Inference Burden to Memory for Personalized Dialogue (2025) · cites this paper as Martins et al., 2022
The retrieved passages have no body-text section or page; they contain metadata and an abstract, which cannot serve as evidence under the stated rubric. Searches for personalized interactions, multiple sessions, and dialogue memory found no eligible passage addressing the claim. (LLM, conf=0.69)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · see citing works
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · see citing works
Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention
Katharopoulos, A., Vyas, A., Pappas, N., & Fleuret, F. (2020) · arXiv · doi:10.48550/arxiv.2006.16236 · Cited by 827 · see citing works
Generalization through Memorization: Nearest Neighbor Language Models
Khandelwal, U., Levy, O., Jurafsky, D., Zettlemoyer, L., & Lewis, M. (2019) · arXiv · doi:10.48550/arxiv.1911.00172 · Cited by 285 · see citing works
Compressive Transformers for Long-Range Sequence Modelling
Rae, J. W., Potapenko, A., Jayakumar, S. M., & Lillicrap, T. P. (2019) · arXiv · doi:10.48550/arxiv.1911.05507 · Cited by 205 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
RULER: What's the Real Context Size of Your Long-Context Language Models?
Hsieh, C.-P., Sun, S., Kriman, S., Acharya, S., Rekesh, D., Jia, F., Zhang, Y., & Ginsburg, B. (2024) · arXiv · Cited by 13 · see citing works
Kim, S., Lee, Y., Kim, S., Kim, H., & Cho, S. (2025) · Findings of the Association for Computational Linguistics: EMNLP 2025 · doi:10.18653/v1/2025.findings-emnlp.1204 · Cited by 2 · see citing works
FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference
Lai, X., Lu, J., Luo, Y., Ma, Y., & Zhou, X. (2025) · arXiv · doi:10.48550/arxiv.2502.20766
Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
Gemini Team, Georgiev, P., Lei, V. I., Burnell, R., Bai, L., Gulati, A., Tanzer, G., Vincent, D., Pan, Z., Wang, S., et al. (2024) · arXiv · doi:10.48550/arxiv.2403.05530 · Cited by 306 · see citing works
- Authors
- Martins, P. H., Marinho, Z., & Martins, A.
- Affiliations
- Instituto de Telecomunicações (PT)
- Venue
- Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
- Published
- 2022
- Pages
- 5468-5485
- ISSN
- 0736-587X (ISSN-L)
- OpenAlex
- W4285272557
- Cited by
- 15
- References
- 5
- Citations by year
- 2026: 1 · 2024: 4 · 2023: 10
- FWCI
- 1.94
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Proceedings of the conference - Association for Computational Linguistics. Meeting (publishedVersion, cc-by)
- Funders
- Horizon 2020 Framework Programme (EU‐H2020)
- Fundação para a Ciência e a Tecnologia (UIDB/50008, UIDB/50008/2020, H2020, 50008, P2020, PTDC/CCI-INF/4703/2021)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
- Keywords
- long-term memory, context length, efficient transformers, language modeling
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
