Metraipolis › archive › dc_d2e607b09f17

ConvMem: Convolutional Memory for Long-Context Reasoning — arXiv:2609.10441

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.10441
License (as archived)
CC0-1.0
Archived
2026-10-16T07:24:45.164Z
SHA-256 of the archived text
25955f5b76ec070bf2fbf7729f1ca88ffc23ebd4a57f9213281bb2b32bc2c53b
Stable URL of this page
https://metraipolis.com/documento/dc_d2e607b09f17

How to cite

ConvMem: Convolutional Memory for Long-Context Reasoning — arXiv:2609.10441 — arXiv (en), https://arxiv.org/abs/2609.10441 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d2e607b09f17 (Metraipolis archive dc_d2e607b09f17, sha256 25955f5b76ec070bf2fbf7729f1ca88ffc23ebd4a57f9213281bb2b32bc2c53b)
BibTeX
@misc{dcd2e607b09f17,
  title = {ConvMem: Convolutional Memory for Long-Context Reasoning — arXiv:2609.10441},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.10441},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d2e607b09f17; sha256 25955f5b76ec070bf2fbf7729f1ca88ffc23ebd4a57f9213281bb2b32bc2c53b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.10441. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

While Large Language Models (LLMs) have demonstrated impressive capabilities, they often struggle with extremely long contexts due to fixed context limits. To address this, sequential approaches like MemAgent extend the effective context by reading text in segments and iteratively updating a fixed-size memory. However, this sequential paradigm suffers from high latency and requires costly reinforcement learning (RL) training, which can lead to overfitting on specific datasets. To overcome these limitations, we propose ConvMem, a training-free, highly parallelizable framework that reformulates long-context reasoning as a hierarchical convolution. Inspired by CNNs, ConvMem treats an LLM prompted with a specific query as a convolutional kernel. This kernel summarizes text segments hierarchically, shortening the reasoning path from a linear chain into a logarithmic tree. Specifically, ConvMem integrates \textit{Configurable Strides} and \textit{Skip Connections} to ensure robust evidence capture and propagation, while employing \textit{Multi-Kernel Convolution} to decompose complex queries into disentangled semantic channels. This design not only mitigates error accumulation but also enables massive parallelization across both text segments and reasoning threads. Experiments on RULER-HotpotQA and RULER-2WikiMultiHopQA demonstrate that ConvMem outperforms training-free baselines and avoids the risk of overfitting to parametric priors often observed in RL-trained models on out-of-distribution tasks.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-10 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.10441 — 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 2026-10-16T07:00:16.258Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-10?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).