The DOI matches its Crossref record; 54 of the 54 references it cites were found in a registry or an index; 21 of the 27 sources its claims cite were checked for support.
HingeMem: Boundary Guided Long-Term Memory with Query Adaptive Retrieval for Scalable Dialogues
Open accessAbstract
Long-term memory is critical for dialogue systems that support continuous, sustainable, and personalized interactions. However, existing methods rely on continuous summarization or OpenIE-based graph construction paired with fixed Top- k retrieval, leading to limited adaptability across query categories and high computational overhead. In this paper, we propose HingeMem, a boundary-guided long-term memory that operationalizes event segmentation theory to build an interpretable indexing interface via boundary-triggered hyperedges over four elements: person, time, location, and topic. When any such element changes, HingeMem draws a boundary and writes the current segment, thereby reducing redundant operations and preserving salient context. To enable robust and efficient retrieval under diverse information needs, HingeMem introduces query-adaptive retrieval mechanisms that jointly decide (a) what to retrieve: determine the query-conditioned routing over the element-indexed memory; (b) how much to retrieve: control the retrieval depth based on the estimated query type. Extensive experiments across LLM scales (from 0.6B to production-tier models; e.g., Qwen3-0.6B to Qwen-Flash) on LOCOMO show that HingeMem achieves approximately 20% relative improvement over strong baselines without query categories specification, while reducing computational cost (68%\downarrow question answering token cost compared to HippoRAG2). Beyond advancing memory modeling, HingeMem's adaptive retrieval makes it a strong fit for web applications requiring efficient and trustworthy memory over extended interactions.
Overview
CoChat Summary
## TL;DR
HingeMem is a neuro-inspired long-term memory system for dialogue that encodes memory at event boundaries and retrieves it with query-adaptive plans rather than fixed Top-k retrieval. On LOCOMO, it improves over strong baselines by about 5% overall, exceeds 10% on multi-hop questions, and reduces question-answering token cost by 68% versus HippoRAG2.
## Key findings
- On LOCOMO, HingeMem achieves the best overall performance across question categories; the paper reports an over 5% improvement in overall metrics and more than 10% on multi-hop questions.
- In the main results table, HingeMem reaches 63.9 F1, 69.4 J, and 0.406 BLEU-1 overall without category-specific question templates; with category format it reaches 75.1 J and 0.404 BLEU-1 overall.
- The ablation table shows a progression from RAG+TM (44.6) to RAG+BM (57.4) to RAG+BM+NI (58.1), then 61.2 with hyperedge reranking and 63.9 with adaptive stop, indicating gains from both boundary-guided memory and query-adaptive retrieval.
- The paper says HingeMem cuts question answering token cost by 68% compared to HippoRAG2 and maintains scalability across LLM sizes from 0.6B to production-tier models.
## Methods
- The system builds boundary-guided long-term memory by detecting dialogue boundaries when key elements change across person, time, location, and topic, then storing boundary-aligned segments as hyperedges linked to element-indexed nodes.
- It evaluates on LOCOMO, which the paper says has 10 total conversations, 27.2 average sessions per conversation, 294.1 average turns, and 15,965.8 average tokens per conversation, with 1,986 total questions across five categories.
- Main analysis uses F1, BLEU-1, and LLM-as-a-Judge, comparing against baselines including RAG, Memorybank, LangMem, HippoRAG2, Mem0, Zep, and others under identical settings.
## Limitations
- The paper states that existing methods can drop 30% to 40% when query categories are not given; this is a limitation of prior work, not HingeMem.
- The text makes evident that evaluation is mainly on one benchmark, LOCOMO, so broader generalization beyond that dataset is not directly shown here.
- The paper notes that some baseline token costs could not be measured accurately because Mem0 and Zep are official services; this is stated in the paper.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
11 Valid · 4 Not valid · 2 Not checked · 80 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“Recent fMRI work further underscores the importance of 'encoding at boundaries,' showing that hippocampal-cortical interactions at these points promote persistent and detailed memory representations [4].”
What the sources say
- Supportedcites [4] [4] Alexander J Barnett, Mitchell Nguyen, James Spargo, Reesha Yadav, Brendan I Cohn-Sheehy, and Charan Ranganath. 2024. Hippocampal-cortical interactions during event boundaries support retention of complex narrative events. Neuron 112, 2 (2024), 319-330. - source
The Discussion reports that hippocampal–posterior medial network connectivity at event offset was associated with subsequent recall and with more details retrieved after a delay. The Abstract specifies that the delay was two days. The finding concerns event offsets in particular, rather than boundaries generally. (LLM, conf=0.90)
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Moreover, most methods employ fixed Topk retrieval strategies ( e.g. Memorybank [56] Top5, Mem0 [7] Top-30), which may introduce both redundancy and noise.”
What the sources say
- Partly supportedcites [56] [56] Wanjun Zhong, Lianghong Guo, Qiqi Gao, He Ye, and Yanlin Wang. 2024. MemoryBank: Enhancing Large Language Models with Long-Term Memory. In AAAI . AAAI Press, 19724-19731. - sourceExperimental — accuracy not yet measured
Section 2.2, “Memory Retrieval” (page not available), confirms that MemoryBank retrieves memory pieces using a dense retrieval model. The passages from our copy do not establish that it uses a fixed Top-5 strategy or discuss redundancy and noise from that strategy; searches for “top-5” and those terms found no supporting passage. (LLM, conf=0.76)
- Partly supportedcites [7] [7] Prateek Chhikara, Dev Khant, Saket Aryan, Taranjeet Singh, and Deshraj Yadav. 2025. Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory. CoRR abs/2504.19413 (2025). - sourceExperimental — accuracy not yet measured
Section 3.3 (p. 8) describes fixed top-k retrieval for the paper’s RAG baseline, with k set to 1 or 2. Section 4.3 (p. 13) says retrieving large chunks can introduce noise. These passages do not establish the attributed Mem0 Top-30 setting; the paper instead describes Mem0’s retrieval as selective in Section 4.4 (p. 13). (LLM, conf=0.84)
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“The effectiveness of such fixedk quickly saturates or even declines for different categories of queries [49].”
What the sources say
- Supportedcites [49] [49] Wujiang Xu, Zujie Liang, Kai Mei, Hang Gao, Juntao Tan, and Yongfeng Zhang. 2025. A-MEM: Agentic Memory for LLM Agents. CoRR abs/2502.12110 (2025). - source
Section 4.5, Hyperparameter Analysis (p. 7), reports that performance across five task categories plateaus and sometimes slightly decreases as the retrieval parameter k increases. Figure 3 likewise describes diminishing gains across categories. This supports the claim about the effectiveness of fixed-k retrieval. (LLM, conf=0.91)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Similarity patterns in the heatmap of vanilla RAG in Figure 1 indicate that fixedk retrieval is suboptimal.”
hedgedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“As illustrated in Figure 1, performance drops around 30% when query categories are unspecified.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
43 more claims
Create a free account to see all 97 claims and to verify the 1 reference still open.
- Probably not a substantial claim. 23
- It may state two separate claims as one. 36
- The quote only partly states this claim. 1
Citations
Cites
Yang, A., Li, A., Yang, B., Zhang, B., Hui, B., Zheng, B., Yu, B., Gao, C., Huang, C., Lv, C., et al. (2025) · arXiv · doi:10.48550/arxiv.2505.09388 · Cited by 17,283
Recognition-by-components: A theory of human image understanding.
Biederman, I. (1987) · Psychological Review · doi:10.1037/0033-295x.94.2.115 · Cited by 5,621
Discovering Event Structure in Continuous Narrative Perception and Memory
Baldassano, C., Chen, J., Zadbood, A., Pillow, J. W., Hasson, U., & Norman, K. A. (2017) · Neuron · doi:10.1016/j.neuron.2017.06.041 · Cited by 952
Perceiving, remembering, and communicating structure in events.
Zacks, J. M., Tversky, B., & Iyer, G. (2001) · Journal of Experimental Psychology: General · doi:10.1037/0096-3445.130.1.29 · Cited by 640
LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory
Wu, D., Wang, H., Yu, W., Zhang, Y., Chang, K.-W., & Yu, D. (2024) · arXiv · doi:10.48550/arxiv.2410.10813 · Cited by 493
Self-citationZacks, J. M., & Swallow, K. M. (2007) · Current Directions in Psychological Science · doi:10.1111/j.1467-8721.2007.00480.x · Cited by 433
MemoryBank: Enhancing Large Language Models with Long-Term Memory
Zhong, W., Guo, L., Gao, Q., Ye, H., & Wang, Y. (2023) · arXiv · doi:10.48550/arxiv.2305.10250 · Cited by 253
In Prospect and Retrospect: Reflective Memory Management for Long-term Personalized Dialogue Agents
Tan, Z., Yan, J., Hsu, I.-H., Han, R., Wang, Z., Le, L., Song, Y., Chen, Y., Palangi, H., Lee, G., et al. (2025) · Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2025.acl-long.413 · Cited by 219
Frames: a corpus for adding memory to goal-oriented dialogue systems
El Asri, L., Schulz, H., Sharma, S., Zumer, J., Harris, J., Fine, E., Mehrotra, R., & Suleman, K. (2017) · Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue · doi:10.18653/v1/w17-5526 · Cited by 212
Structured Event Memory: A neuro-symbolic model of event cognition.
Franklin, N. T., Norman, K. A., Ranganath, C., Zacks, J. M., & Gershman, S. J. (2020) · Psychological Review · doi:10.1037/rev0000177 · Cited by 177
Research works printed but not linked: 5
- Institute for Basic Science. 2023. AI's memory-forming mechanism found to be strikingly similar to that of the brain. ScienceDaily (2023).
- Yubin Ge, Salvatore Romeo, Jason Cai, Raphael Shu, Yassine Benajiba, Monica Sunkara, and Yi Zhang. 2025. TReMu: Towards Neuro-Symbolic Temporal Reasoning for LLM-Agents with Memory in Multi-Session Dialogues. In ACL (Findings). Association for Computational Linguistics, 18974-18988.
- Junqing He, Liang Zhu, Rui Wang, Xi Wang, Gholamreza Haffari, and Jiaxing Zhang. 2025. MADial-Bench: Towards Real-world Evaluation of Memory-Augmented Dialogue Generation. In NAACL (Long Papers). Association for Computational Linguistics, 9902-9921.
- Qihong Lu, Uri Hasson, and Kenneth A Norman. 2022. A neural network model of when to retrieve and encode episodic memories. elife, Vol. 11 (2022), e74445.
- Jing Xu, Arthur Szlam, and Jason Weston. 2022b. Beyond Goldfish Memory: Long-Term Open-Domain Conversation. In ACL (1). Association for Computational Linguistics, 5180-5197.
Cited by
C 3 Flow: SIMD-Style Concurrent Claude Code Workflow for Scaling Deep Research
Gao, Y., Huang, X., Zhong, Y., Fan, Z., Gui, Z., Liang, L., Xiong, Y., & Wang, H. (2026) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3805712.3808374
Self-citation
Metadata
- Authors
- Zhong, Y., Gao, Y., & Wang, H.
- Affiliations
- Tongji University (CN)
- Venue
- Proceedings of the ACM Web Conference 2026
- Publisher
- ACM
- Published
- 2026-04-12
- Pages
- 3576-3587
- OpenAlex
- W7152670197
- Cited by
- 1
- References
- 8
- Citations by year
- 2026: 1
- FWCI
- 8.94
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- International World Wide Web Conference (WWW) (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Funders
- the National Natural Science Foundation of China (U23B2057)
- Shanghai Pilot Program for Basic Research (22TQ1400300)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech and dialogue systems
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
- Keywords
- long-term memory, dialogue systems, event segmentation, query routing, question answering, computational efficiency
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: HingeMem: Boundary Guided Long-Term Memory with Query Adaptive Retrieval for Scalable Dialogues. https://app.cochat.ai/papers/10.1145/3774904.3792089
- The paper
- Zhong, Y., Gao, Y., & Wang, H. (2026). HingeMem: Boundary Guided Long-Term Memory with Query Adaptive Retrieval for Scalable Dialogues. Proceedings of the ACM Web Conference 2026. https://doi.org/10.1145/3774904.3792089
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.