BlackArc10 papers
Memory Systems
How agents retain, recall, and reflect across long-running work.
- Papers
- 10
- Read time
- 2.7h
- Avg length
- 16m
HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models
Use an LLM to turn your whole corpus into one big knowledge graph, then answer multi-hop questions in a *single* retrieval step by running Personalized PageRank from the query's entities — instead of paying for 3-5 round trips of iterative LLM-driven retrieval.
- Read
- 16 min
- Source
- arXiv:2405.14831
MemGPT: Towards LLMs as Operating Systems
Treat the LLM's context window like RAM and give the model OS-style function calls to page its own memory in and out of an external database, so a fixed-context model behaves as if it had unbounded memory.
- Read
- 14 min
- Source
- arXiv:2310.08560
From RAG to Memory: Non-Parametric Continual Learning for Large Language Models
HippoRAG 2 turns a pile of documents into a queryable "long-term memory" by building a knowledge graph over passages *and* concepts, then using Personalized PageRank to retrieve — beating plain vector search on multi-hop questions without breaking on simple ones.
- Read
- 14 min
- Source
- arXiv:2502.14802
Memento: Fine-tuning LLM Agents without Fine-tuning LLMs
Instead of retraining the LLM, Memento gives an agent a growing "case bank" of its own past successes and failures and learns a tiny side-model that decides which past cases to recall — so the agent keeps getting better on the job without ever touching the model weights.
- Read
- 16 min
- Source
- arXiv:2508.16153
ArcMemo: Abstract Reasoning Composition with Lifelong LLM Memory
Instead of caching whole question/answer pairs, ArcMemo distills each solved problem into small, reusable, *named* concepts in plain language, then selectively pulls the relevant ones into the prompt for the next problem — giving an LLM a growing, composable skill library without any fine-tuning.
- Read
- 16 min
- Source
- arXiv:2509.04439
FERNme — Action-Coupled, Cost-Bounded Memory for Multi-Tenant Agents
Stop paying an LLM to "remember" each user on every turn — instead keep each user
- Read
- 11 min
Filesystem-Based Memory for LLM Agents
The first systematic test of the memory design every serious agent product has quietly converged on — a folder of markdown files the agent edits with its own file tools — and it finds that letting the agent organize that folder cuts search cost but does not, on its own, produce better answers.
- Read
- 23 min
Hindsight is 20/20: Building Agent Memory that Retains, Recalls, and Reflects
A memory architecture that stops treating agent memory as a dumb top-k vector store and instead splits it into four "networks" (facts vs. experiences vs. opinions vs. summaries) with three operations (retain / recall / reflect), lifting a 20B open model from 39% to 83.6% on long-horizon memory benchmarks — beating full-context GPT-4o.
- Read
- 18 min
- Source
- arXiv:2512.12818
MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents
Instead of letting an agent's context balloon as it works through a long, multi-step task, MEM1 trains the model with RL to rewrite a single compact "internal state" each turn — folding new observations into it and throwing everything else away — so memory stays roughly constant no matter how long the task runs.
- Read
- 16 min
- Source
- arXiv:2506.15841
MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent
Instead of stretching a model's context window, MemAgent teaches an LLM (via RL) to read a giant document in small chunks while continuously rewriting a fixed-size scratchpad of notes — so an 8K-window model trained on 32K text answers questions over 3.5M-token documents with under 5% accuracy loss and linear cost.
- Read
- 18 min
- Source
- arXiv:2507.02259