Metraipolis › archive › dc_a072884d689d

MEMO: Multimodal Evidence Memory Organization for Long-Horizon LLM Agents — arXiv:2609.07471

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.07471
License (as archived)
CC0-1.0
Archived
2026-10-16T12:45:37.058Z
SHA-256 of the archived text
3699e098dbde25427e20127be69d34dd1f6f91f16bc4d89a47c048a968e2621a
Stable URL of this page
https://metraipolis.com/documento/dc_a072884d689d

How to cite

MEMO: Multimodal Evidence Memory Organization for Long-Horizon LLM Agents — arXiv:2609.07471 — arXiv (en), https://arxiv.org/abs/2609.07471 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a072884d689d (Metraipolis archive dc_a072884d689d, sha256 3699e098dbde25427e20127be69d34dd1f6f91f16bc4d89a47c048a968e2621a)
BibTeX
@misc{dca072884d689d,
  title = {MEMO: Multimodal Evidence Memory Organization for Long-Horizon LLM Agents — arXiv:2609.07471},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.07471},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a072884d689d; sha256 3699e098dbde25427e20127be69d34dd1f6f91f16bc4d89a47c048a968e2621a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.07471. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Long-running LLM agents rely on external memory to store and reuse information beyond a single context window, yet there is a fundamental tension between the continuous accumulation of interaction trajectories and the limited context capacity. The key challenge in agent memory is therefore not only to retrieve relevant records, but also to select necessary evidence under a given budget and organize it in an appropriate modality. Existing memory readout methods mainly use textual or visual forms. Text preserves high fidelity, but its linear token representation makes contents with different importance compete for the limited context at nearly uniform unit cost. Visual readout renders text into document-like images, which can use two-dimensional layouts to expose structure and emphasize key information, but it may lose fine-grained details during rendering and compression. To address this issue, we propose MEMO, a multimodal evidence memory organization method for LLM agents. MEMO first uses a trained evidence extractor to select relevant memory blocks and form evidence units with source information and presentation requirements. A trained query-conditioned memory manager assigns each unit to a textual, visual, or dual-channel carrier and selects a layout that matches the evidence structure. A deterministic memory construction module then generates the textual package and visual pages. The memory manager is trained with feedback from an offline reader that measures the utility of the guided memory plan, so that retention and presentation decisions align with downstream usage. We evaluate MEMO on four benchmarks, HotpotQA, 2WikiMultiHopQA, LoCoMo, and ALFWorld, with multiple reader backends. The results show that MEMO presents memory more efficiently with fewer memory tokens, improves downstream task performance, and builds more effective working memory under constrained budgets.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.07471 — 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 2026-10-16T12:30:25.121Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).