Trajectory Memory Compression for Long-Horizon Dialogue Agents with Provenance Preservation
Keywords:
Trajectory Memory, Memory Compression, Long-Horizon Dialogue, Provenance Preservation, Autonomous AgentsAbstract
The rapid advancement of artificial intelligence has facilitated the development of dialogue agents capable of highly sophisticated interactions. However, sustaining coherence and factual consistency over extended interactions remains a formidable challenge due to the inherent constraints of context windows in large language models. This paper investigates the critical problem of trajectory memory compression in long-horizon dialogue agents, introducing a novel paradigm that integrates provenance preservation into the compression pipeline. Traditional memory compression techniques often discard granular contextual details, leading to hallucinations and a loss of traceability when the agent attempts to recall historical information. By conceptualizing dialogue history as a continuous trajectory of events, we propose a methodology that selectively compresses semantic content while meticulously maintaining cryptographic and structural pointers to the original episodic occurrences. This dual-track approach ensures that the agent can retrieve highly compressed semantic representations without sacrificing the ability to ground its responses in specific past utterances. Through rigorous theoretical analysis and extensive empirical evaluation across multiple long-context dialogue datasets, we demonstrate that our provenance-aware compression mechanism significantly enhances both the computational efficiency of memory retrieval and the factual reliability of the generated responses. The findings offer a scalable framework for deploying autonomous agents in domains requiring prolonged interaction and strict auditing, such as healthcare and legal counseling.
References
1. Zhang, B., Wang, S., Jiang, Y., Sui, D., Tu, Z., & Chu, D. (2025, July). Ask and Retrieve Knowledge: Towards Proactive Asking with Imperfect Information in Medical Multi-turn Dialogues. In Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 1055-1065).
2. Z. Wang, R. Yuan, Z. Geng, H. Li, X. Qu, X. Li, S. Chen,
H. Fu, R. B. Dannenberg, and K. Zhang, ‘‘Singing timbre popularity
assessment based on multimodal large foundation model,’’ in Proc.
33rd ACM Int. Conf. Multimedia, Oct. 2025, pp. 12227–12236, doi:
10.1145/3746027.3758148.
3. Yao, Z., Li, R., Chen, X., Cheng, J., Le, C., Yue, M., ... & Dong, J. (2026). PreDiff-LM: Pretrained Discrete Masked Diffusion Language Modeling with Hybrid Attention. arXiv preprint arXiv:2607.25157.
4. Liu, X., Yu, Z., Yao, Y., Yang, K., Zeng, H., & Chen, C. P. (2026). A Comprehensive Survey of Cross-Modal Retrieval: Breaking Through Modal Barriers. ACM Computing Surveys.
5. Yu, H., Zhang, J., Chen, C., Xiang, T., Fang, Y., Niebles, J. C., & Adeli, E. (2026, March). Socialgen: Modeling multi-human social interaction with language models. In 2026 International Conference on 3D Vision (3DV) (pp. 1-17). IEEE.
6. Wang, Z., Hou, J., & Zhang, H. (2026, July). Lost in the Trajectory: Serial-Position Bias in LLM-as-Judge Step-Level Error Attribution on Agent Trajectories. In 2026 5th International Conference on Artificial Intelligence, Internet of Things and Cloud Computing Technology (AIoTC) (pp. 354-358). IEEE.
7. Zou, Y., Tong, J., & Li, J. (2026). Text-Preserving Lossy Text Compression: A Study of Strategic Deletion and LLM Reconstruction. arXiv preprint arXiv:2605.29000.
8. Peters, M. E., Neumann, M., Iyyer, M., Gardner, M., Clark, C., Lee, K., & Zettlemoyer, L. (2018). Deep contextualized word representations. In Proceedings of NAACL-HLT.
9. Lyu, S., Deng, Y., Liu, G., Qi, Z., & Wang, R. (2025, June). Transferable Modeling Strategies for Low-Resource LLM Tasks: A Prompt and Alignment-Based Approach. In 2025 7th International Conference on Artificial Intelligence Technologies and Applications (ICAITA) (pp. 225-229). IEEE.
10. Yao, Z., Li, R., Chen, X., Cheng, J., Le, C., Yue, M., ... & Dong, J. (2026). Less Data, Better Alignment: Data-Centric Multi-Evaluator Agreement for Preference Optimization. arXiv preprint arXiv:2607.25136.
11. Yao, Y., Xiao, W., Miao, P., Chen, G., Yang, H., Chae, C. B., & Wong, K. K. (2026). UAV-RHS-enabled full-duplex ISAC covert system: Robust beamforming and trajectory optimization. IEEE Transactions on Communications.
12. Pei, Z., Zhang, A., Wang, S., & Huang, Q. (2024, July). Modeling language tokens as functionals of semantic fields. In Forty-first International Conference on Machine Learning.
13. Zhu, R., Ma, Z., Wu, J., Gao, J., Wang, J., Lin, D., & He, C. (2025, April). Utilize the flow before stepping into the same river twice: Certainty represented knowledge flow for refusal-aware instruction tuning. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 39, No. 24, pp. 26157-26165).
14. Sun, J., Lin, Y., Xue, Y., Wang, Y., Yao, Z., Qian, R., ... & Zhou, H. (2026). Pmmc: Prospective multimodal memory compilation for long-term lvlm agents. arXiv preprint arXiv:2608.00962.
15. Yang, H., Xie, Z., Xiong, S., & Huang, B. (2026). DepthBenchCAD: When Does Deeper Auditing Yield More Reliable Conclusions?. arXiv preprint arXiv:2609.15122.
16. Fensham, R., Sumner, T. D., Cutter, N., Buchanan, G., Liu, R., Munoz, J., ... & Van Der Plaat, D. (2024). Towards a National Data Architecture for Cultural Collections: Designing the Australian Cultural Data Engine. Digital Humanities Quarterly, 18(2).
17. Liu, R., Zhang, R., & Wang, S. (2025, December). Transformer-based modeling of user interaction sequences for dwell time prediction in human-computer interfaces. In 2025 7th International Academic Exchange Conference on Science and Technology Innovation (IAECST) (pp. 352-356). IEEE.
18. Liu, H., & Henao, R. (2025, April). Learning to substitute words with model-based score ranking. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) (pp. 11551-11565).
19. Wang, Y., Fan, S., Wang, M., Gao, S., & Wang, C. (2026, April). DAMR: Efficient and adaptive context-aware knowledge graph question answering with LLM-guided MCTS. In International Conference on Learning Representations (Vol. 2026, pp. 25548-25570).
20. Hu, Y., Yang, Z., Zhao, C., & Ji, W. (2026, May). Adaptive guidance semantically enhanced via multimodal llm for edge-cloud object detection. In ICASSP 2026-2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 12962-12966). IEEE.
21. Papineni, K., Roukos, S., Ward, T., & Zhu, W.-J. (2002). BLEU: A method for automatic evaluation of machine translation. In Proceedings of ACL.
22. Sun, J., Xue, Y., Li, M., Yao, Z., Li, J., Cui, Y., ... & Zhou, H. (2026). TrajWiki: Source-Grounded Memory Trajectories for Long-Horizon Dialogue Agents. arXiv preprint arXiv:2608.00967.
23. Wu, J., Chen, S., Heo, I., Gutfraind, S., Liu, S., Li, C., ... & Sharps, M. (2025). Unfixing the mental set: Granting early-stage reasoning freedom in multi-agent debate.
24. Li, Y. (2026, July). Retrieval-Augmented Generation for Historical Fault Experience Reuse and Intelligent Diagnosis. In 2026 International Conference on Large Models and Intelligent Agent Technologies (LMIAT) (pp. 47-51). IEEE.
25. Yao, Y., Xiao, W., Miao, P., Chen, G., Yang, H., Chae, C. B., & Wong, K. K. (2025). UAV-relay-aided secure maritime networks coexisting with satellite networks: robust beamforming and trajectory optimization. IEEE Transactions on Wireless Communications.
26. Liu, J., Yang, X., Wang, C., & Tao, Y. (2018, September). A convolution neural network for dolphin species identification using echolocation clicks signal. In 2018 IEEE International Conference on Signal Processing, Communications and Computing (ICSPCC) (pp. 1-4). IEEE.
27. Lyu, S., Mao, S., & Ma, Y. (2025). Unified Long Video Inpainting and Outpainting via Overlapping High-Order Co-Denoising. arXiv preprint arXiv:2511.03272.
28. Chen, S., Zhu, T., Wang, Z., Zhang, J., Wang, K., Zhou, R., ... & Li, M. (2025). Why Do LLM Agents Fail in Exploring New Environments? A World-Modeling Perspective. arXiv preprint arXiv:2510.15047.
29. Fang, S., Shen, I., Wang, Y., Tsai, Y. H., Yang, Y., Zhou, S., ... & Yang, M. H. (2025). Meshllm: Empowering large language models to progressively understand and generate 3d mesh. In Proceedings of the IEEE/CVF international conference on computer vision (pp. 14061-14072).
30. Zhang, H., & Wang, Z. (2026). ProvLevel-Reason: An Oracle-Scored Benchmark for LLM Reasoning over Software Build-Provenance Attestations. International Journal of Information and Communication Technology Trends, 6(1), 269-276.
31. Banerjee, S., & Lavie, A. (2005). METEOR: An automatic metric for MT evaluation with improved correlation with human judgments. In Proceedings of the ACL Workshop on Intrinsic and Extrinsic Evaluation Measures for MT and/or Summarization.
32. Li, Y. (2026, June). AI-augmented automated triage for system error reporting using log and memory dump analysis. In 2026 7th International Conference on Big Data & Artificial Intelligence & Software Engineering (ICBASE) (pp. 572-577). IEEE.
33. Mo, Z. (2025, May). Development of an Intelligent Retrieval and Recommendation System for Chinese Language Educational Resources Based on DNN Technology. In 2025 2nd International Conference on Intelligent Computing and Robotics (ICICR) (pp. 170-175). IEEE.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Philip Ho, Joanna Ma
This work is licensed under a Creative Commons Attribution 4.0 International License.
All articles published in the Journal of Social Sciences, Humanities and Innovation (JSSHI) are licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
This license permits unrestricted use, distribution, reproduction, and adaptation in any medium or format, provided that the original author(s) and source are properly credited. Users are allowed to copy, redistribute, remix, transform, and build upon the material for any purpose, including commercial use, as long as appropriate attribution is given and any changes made are indicated.
Authors retain copyright of their work while granting the journal the right of first publication. The journal supports open access principles and encourages the wide dissemination of scholarly research to promote knowledge sharing and academic collaboration.
For more information about the license, visit: https://creativecommons.org/licenses/by/4.0/
