Paper TL;DR

Research papers on AI memory, retrieval and search, each boiled down to what it says, why it matters, and where it falls short. 12 so far · RSS for this section

1. Six Papers That Shaped How I Think About AI Memory

   Short, plain-language summaries of six papers behind my AI memory side projects — from Hopfield's 1982 associative memory to user-owned memory for agents.

   Oct 18, 2026 3 min read
2. Hopfield Networks Is All You Need — Attention Is a Memory Lookup

   A plain-language walkthrough of Ramsauer et al. (2020): how modern Hopfield networks store exponentially many memories and why their update rule is transformer attention — with 20 lines of NumPy.

   Oct 18, 2026 3 min read
3. Complementary Learning Systems — Why Agents Need Two Kinds of Memory

   A plain-language TL;DR of McClelland, McNaughton & O'Reilly (1995): the brain's fast and slow learners, catastrophic interference, and what it suggests for designing memory in AI agents.

   Oct 18, 2026 3 min read
4. LoCoMo — How Good Is Your Agent's Memory, Really?

   A plain-language TL;DR of the LoCoMo benchmark (Maharana et al., 2024) for very long-term conversational memory — what it tests, what it found, and how I used it to evaluate VaultMem.

   Oct 18, 2026 2 min read
5. Memory as Asset — Whose Memory Is It Anyway?

   A plain-language TL;DR of Memory as Asset (Pan, Huang & Yang, 2026): moving AI memory from agent-centric to human-centric, its three principles and three layers, and how it shaped VaultMem.

   Oct 18, 2026 2 min read
6. Sentence-BERT — The Quiet Workhorse Behind Semantic Search

   A plain-language TL;DR of Sentence-BERT (Reimers & Gurevych, 2019): why plain BERT is too slow for search, how siamese training fixes it, and why it still powers most semantic search.

   Oct 18, 2026 2 min read
7. Hopfield (1982) — Memory as a Valley You Roll Into

   A plain-language TL;DR of John Hopfield's 1982 paper on associative memory: energy landscapes, Hebbian storage, recall from partial cues and its limits — with a 25-line NumPy demo.

   Oct 18, 2026 2 min read
8. MemGPT — Treating the Context Window Like RAM

   A plain-language TL;DR of MemGPT (Packer et al., 2023): how an LLM can page information in and out of a small context window with function calls, the way an operating system manages memory.

   Oct 19, 2026 4 min read
9. Generative Agents — Memory Streams, Reflection and Plans

   A plain-language TL;DR of Generative Agents (Park et al., 2023): how 25 LLM characters in a sandbox town remember, reflect and plan, and which parts of that memory design builders still copy.

   Oct 19, 2026 5 min read
10. Lost in the Middle — Where You Put Context Matters

    A plain-language TL;DR of Lost in the Middle (Liu et al., 2023): language models use information at the start and end of a long prompt far better than information buried in the middle, and what that means for RAG and agent memory.

    Oct 20, 2026 4 min read
11. RAG — The Paper That Named Retrieval-Augmented Generation

    A plain-language TL;DR of Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks (Lewis et al., 2020): a language model paired with a searchable Wikipedia index, trained together, and why you can update what it knows by swapping the index.

    Oct 20, 2026 5 min read
12. HNSW — How Vector Search Skips Most of the Haystack

    A plain-language TL;DR of Hierarchical Navigable Small World graphs (Malkov & Yashunin): a layered neighbor graph that finds approximate nearest vectors in roughly logarithmic time, and the index behind a lot of today's vector search.

    Oct 20, 2026 5 min read

The papers themselves, with one-line notes, are on the papershelf.