# $δ$-mem: Efficient Online Memory for Large Language Models

Page: https://stenobird.com/podcast/daily-paper-cast-7079649/mem-efficient-online-memory-for-large-language-models
Text version: https://stenobird.com/podcast/daily-paper-cast-7079649/mem-efficient-online-memory-for-large-language-models.md
Podcast: [Daily Paper Cast](https://stenobird.com/podcast/daily-paper-cast-7079649)
Published: 2026-05-14T04:33:18+00:00
Episode link: https://share.transistor.fm/s/979c7e38
Audio file: https://media.transistor.fm/979c7e38/96fc0601.mp3
Processing state: not_requested
JSON: https://stenobird.com/v1/public/podcasts/daily-paper-cast-7079649/episodes/mem-efficient-online-memory-for-large-language-models
Duration seconds: 1471

## Resource

🤗 Upvotes: 90 | cs.AI Authors: Jingdi Lei, Di Zhang, Junxian Li, Weida Wang, Kaixuan Fan, Xiang Liu, Qihan Liu, Xiaoteng Ma, Baian Chen, Soujanya Poria Title: $δ$-mem: Efficient Online Memory for Large Language Models Arxiv: http://arxiv.org/abs/2605.12357v1 Abstract: Large language models increasingly need to accumulate and reuse historical information in long-term assistants and agent systems. Simply expanding the context window is costly and often fails to ensure effective context utilization. We propose $δ$-mem, a lightweight memory mechanism that augments a frozen full-attention backbone with a compact online state of associative memory. $δ$-mem compresses past information into a fixed-size state matrix updated by delta-rule learning, and uses its readout to generate low-rank corrections to the backbone's attention computation during generation. With only an $8\times8$ online memory state, $δ$-mem improves the average score to $1.10\times$ that of the frozen backbone and $1.15\times$ that of the strongest non-$δ$-mem memory baseline. It achieves larger gains on memory-heavy benchmarks, reaching $1.31\times$ on MemoryAgentBench and $1.20\times$ on LoCoMo, while largely preserving general capabilities. These results show that effective memory can be realized through a compact online state directly coupled with attention computation, without full fine-tuning, backbone replacement, or explicit context extension.

## Actions

- request_transcript: `POST https://stenobird.com/v1/public/podcasts/daily-paper-cast-7079649/episodes/mem-efficient-online-memory-for-large-language-models/transcription-requests` — Idempotently request low-priority transcript generation for this episode.
- read_markdown: `GET https://stenobird.com/podcast/daily-paper-cast-7079649/mem-efficient-online-memory-for-large-language-models.md` — Read the agent-friendly Markdown representation of this episode resource.

A page view does not enqueue transcription. Agents should invoke `request_transcript` explicitly when they need this episode processed.

## Transcript

Full transcripts are not published on public pages unless there is a clear rights basis.