Metraipolis › archive › dc_ffb07272f75f

RippleMem: From Isolated Retrieval to Associative Recollection for Long-Term Agent Memory — arXiv:2608.13334

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.13334
License (as archived)
CC0-1.0
Archived
2026-10-19T04:24:38.038Z
SHA-256 of the archived text
dd210694bd637960065ed9abc0514f949eaf073db20f45f8c67bc9ed0db86119
Stable URL of this page
https://metraipolis.com/documento/dc_ffb07272f75f

How to cite

RippleMem: From Isolated Retrieval to Associative Recollection for Long-Term Agent Memory — arXiv:2608.13334 — arXiv (en), https://arxiv.org/abs/2608.13334 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ffb07272f75f (Metraipolis archive dc_ffb07272f75f, sha256 dd210694bd637960065ed9abc0514f949eaf073db20f45f8c67bc9ed0db86119)
BibTeX
@misc{dcffb07272f75f,
  title = {RippleMem: From Isolated Retrieval to Associative Recollection for Long-Term Agent Memory — arXiv:2608.13334},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.13334},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ffb07272f75f; sha256 dd210694bd637960065ed9abc0514f949eaf073db20f45f8c67bc9ed0db86119},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.13334. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LLM-based agents increasingly rely on external memory to support long-horizon reasoning and interaction. However, the main bottleneck is not simply storing past experience, but recovering the right set of evidence when relevant information is distributed across many interactions. Existing approaches struggle with this access problem. Full-context methods require noisy long-context search, flat retrieval often returns isolated and incomplete records, and graph-based memory systems can be expensive to construct while compressing rich event context. We introduce RippleMem, a long-term memory system that replaces one-shot retrieval with adaptive associative recollection. Inspired by cue-dependent episodic retrieval and associative completion, RippleMem stores interaction history as cue-rich episodic memory units and organizes them in an event-centric memory graph. Given a query, it first recalls relevant memory anchors through hybrid cues, then expands from these anchors along semantic and structural associations to recover missing supporting evidence. In this way, initially recalled memories serve not only as answer context, but also as cues for completing the evidence needed to answer. Experiments on LoCoMo and LongMemEval-S show that RippleMem achieves the best overall performance across evaluated settings, improving LLM-as-a-Judge accuracy by 3.95% on LoCoMo and up to 11.87% on LongMemEval-S, while reducing graph construction cost by about 30x.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-14 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.13334 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T04:17:42.398Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).