# Sources and further reading

Primary references for the memory story, representations, and evaluation.

Author: Murali Sid (https://linkedin.com/in/muralisid)
Source: https://www.agenticarchitectureskills.com/memory/sources
Published: 2026-09-13
Updated: 2026-10-15
Licence: CC BY-SA 4.0, https://creativecommons.org/licenses/by-sa/4.0/
Cite as: Murali Sid. Sources and further reading. Agentic Architecture Skills. https://www.agenticarchitectureskills.com/memory/sources

---


The main pages cite sources beside the relevant ideas. This list supports deeper reading. Research findings apply to the authors' tested settings; documentation describes a particular implementation. Reviewed 18 September 2026.

## Memory roles and context [#memory-roles-and-context]

* [CoALA](https://arxiv.org/abs/2309.02427), 2023, revised 2024: a framework for language-agent memory and actions.
* [LangGraph memory overview](https://docs.langchain.com/oss/python/concepts/memory), implementation documentation: thread-scoped state and cross-thread memory.
* [MemGPT](https://arxiv.org/abs/2310.08560), 2023: managing information across memory levels.
* [Generative Agents](https://arxiv.org/abs/2304.03442), 2023: memories, reflections, and planning in a simulated environment, not enterprise validation.

## Representations and storage [#representations-and-storage]

* [Sentence-BERT](https://arxiv.org/abs/1908.10084), 2019: sentence representations.
* [CLIP](https://openai.com/index/clip/), 2021: aligned image and text representations.
* [InternVideo2](https://arxiv.org/abs/2403.15377), 2024: video representation research.
* [CLAP](https://arxiv.org/abs/2206.04769), 2022: audio and language representations.
* [Google Satellite Embedding dataset documentation](https://developers.google.com/earth-engine/datasets/catalog/GOOGLE_SATELLITE_EMBEDDING_V1_ANNUAL), built with Google DeepMind's AlphaEarth Foundations model: annual satellite embeddings and their interpretation.
* [TS2Vec](https://arxiv.org/abs/2106.10466), 2021: time-series representations.
* [RAPTOR](https://arxiv.org/abs/2401.18059), 2024: tree-organized summaries for retrieval.
* [Titans](https://arxiv.org/abs/2501.00663), submitted December 2024: neural memory updated at test time.

## Evaluation and scale [#evaluation-and-scale]

* [LoCoMo](https://arxiv.org/abs/2402.17753), 2024: conversational memory evaluation.
* [LongMemEval](https://arxiv.org/abs/2410.10813), first published 2024: long-term interactive memory abilities.
* [BEAM](https://arxiv.org/abs/2510.27246), 2025, revised 2026: generated conversations reaching ten million tokens.
* [NoLiMa](https://arxiv.org/abs/2502.05167), 2025: long-context evaluation beyond literal matching.
* [MemPalace benchmark notes](https://github.com/MemPalace/mempalace/blob/main/benchmarks/BENCHMARKS.md), project-published: metric and tuning disclosures. These do not independently validate the guide author's reported experiments.

## Operational design [#operational-design]

* [NVIDIA video-search blueprint](https://build.nvidia.com/nvidia/video-search-and-summarization/blueprintcard), vendor documentation.
* [Roboflow Workflows](https://docs.roboflow.com/workflows), vendor documentation.
* [NIST AI RMF Playbook](https://www.nist.gov/itl/ai-risk-management-framework/nist-ai-rmf-playbook), lifecycle and governance guidance.

Provider prices, current product availability, and benchmark leaderboards require fresh verification for a procurement decision. They are not needed to understand the memory foundations.

**Next step:** see how memory learnings are promoted under control on [the learning flywheel](/architecture/learning-flywheel), or return to [Start here](/start).