Category index
All memory research categories
Browse AI and agent memory research by category — agent memory, RAG, long-term context, benchmarks, architectures, and more — with plain-language explainers.
16 categories
Agent Memory
Papers on memory for LLM agents: recall, reflection, and long-horizon context.
118 papers
Benchmark
Empirical studies and benchmarks on context, recall, and memory limitations in LLMs.
31 papers
Cognitive Architecture
Cognitive science–inspired frameworks for language agents — dual-process systems, neuro-symbolic memory, and biomimetic architectures.
8 papers
Continual Learning
Continual and lifelong learning with memory — catastrophic forgetting, memory consolidation, and evolving agent memory.
17 papers
Episodic Memory
Episodic memory in AI agents — storing, retrieving, and reasoning over past experiences and events.
35 papers
Graph Memory
Graph-based memory for AI agents — knowledge graphs, hypergraph retrieval, and relational memory structures.
14 papers
KV Cache
KV cache research — key-value cache management, compression, injection, and efficient serving for LLMs.
13 papers
Long-Term Memory
Long-term and persistent memory mechanisms for conversational AI and assistants.
85 papers
Memory Architecture
Architectures and systems for organizing, storing, and accessing AI memory.
71 papers
Personalization
Personalized memory for AI assistants — user preference learning, persona modeling, and adaptive long-horizon dialogue.
29 papers
Procedural Memory
Procedural memory in LLM agents — learning skills, rules, and how-to knowledge from experience.
12 papers
RAG
Research on retrieval-augmented generation (RAG) and non-parametric memory for language models.
47 papers
Security & Privacy
Security and privacy of AI memory systems — memory attacks, data extraction, privacy-aware agents, and memory governance.
9 papers
Semantic Memory
Semantic memory for LLMs — structured knowledge, fact storage, and meaning-based retrieval.
8 papers
Survey
Survey papers and overview articles on AI memory, agents, and retrieval — explained in plain language.
8 papers
Working Memory
Working memory and short-term context management in LLMs — context windows, compression, and scratchpad mechanisms.
27 papers