Library · 19-retrieval-augmentation-context-management
RAG & context mgmt
| Title | Peer | Link |
|---|---|---|
| RETRO — Improving language models by retrieving from trillions of tokens | ◦ preprint | arxiv.org/abs/2112.04426 |
| Atlas — Few-shot Learning with Retrieval Augmented Language Models | ◦ preprint | arxiv.org/abs/2208.03299 |
| REPLUG — Retrieval-Augmented Black-Box Language Models | ◦ preprint | arxiv.org/abs/2301.12652 |
| IRCoT — Interleaving Retrieval with Chain-of-Thought for multi-step QA | ◦ preprint | arxiv.org/abs/2212.10509 |
| Self-RAG — Learning to Retrieve, Generate and Critique through Self-Reflection | ◦ preprint | arxiv.org/abs/2310.11511 |
| FLARE — Active Retrieval Augmented Generation | ◦ preprint | arxiv.org/abs/2305.06983 |
| Adaptive-RAG — adapting retrieval to question complexity | ◦ preprint | arxiv.org/abs/2403.14403 |
| Adaptive-k — Efficient Context Selection for Long-Context QA | ◦ preprint | arxiv.org/abs/2506.08479 |
| UAR — Unified Active Retrieval for RAG | ◦ preprint | arxiv.org/abs/2406.12534 |
| Self-Route — RAG or Long-Context LLMs? (hybrid routing) | ◦ preprint | arxiv.org/abs/2407.16833 |
| DyCP — Dynamic Context Pruning for Long-Form Dialogue | ◦ preprint | arxiv.org/abs/2601.07994 |
| Knowing When to Stop — Efficient Context Processing via Latent Sufficiency Signals | ◦ preprint | arxiv.org/abs/2502.01025 |
| Quest — Query-Aware Sparsity for Efficient Long-Context Inference | ◦ preprint | arxiv.org/abs/2406.10774 |
| SnapKV — LLM Knows What You Are Looking For Before Generation | ◦ preprint | arxiv.org/abs/2404.14469 |
| PyramidKV — Dynamic KV Cache Compression via Pyramidal Information Funneling | ◦ preprint | arxiv.org/abs/2406.02069 |
| Ada-KV — Adaptive Budget Allocation for KV Cache Eviction | ◦ preprint | arxiv.org/abs/2407.11550 |
| DynamicKV — Task-Aware Adaptive KV Cache Compression | ◦ preprint | arxiv.org/abs/2412.14838 |
| TokenSelect — Dynamic Token-Level KV Cache Selection | ◦ preprint | arxiv.org/abs/2411.02886 |
| FastGen — Model Tells You What to Discard: Adaptive KV Cache Compression | ◦ preprint | arxiv.org/abs/2310.01801 |
| Scissorhands — Persistence of Importance for KV Cache Compression at Test Time | ◦ preprint | arxiv.org/abs/2305.17118 |
| LLMLingua — Compressing Prompts for Accelerated Inference | ◦ preprint | arxiv.org/abs/2310.05736 |
| LongLLMLingua — Prompt Compression for Long-Context Scenarios | ◦ preprint | arxiv.org/abs/2310.06839 |
| AutoCompressors — Adapting Language Models to Compress Contexts | ◦ preprint | arxiv.org/abs/2305.14788 |
| ACON — Optimizing Context Compression for Long-horizon LLM Agents | ◦ preprint | arxiv.org/abs/2510.00615 |
| Infini-attention — Leave No Context Behind: Efficient Infinite Context Transformers | ◦ preprint | arxiv.org/abs/2404.07143 |
| Compressive Transformer — Compressive Transformers for Long-Range Sequence Modelling | ✓ peer | arxiv.org/abs/1911.05507 |
| MemGPT — Towards LLMs as Operating Systems | ◦ preprint | arxiv.org/abs/2310.08560 |
| LongMem — Augmenting Language Models with Long-Term Memory | ◦ preprint | arxiv.org/abs/2306.07174 |
| MemoryBank — Enhancing LLMs with Long-Term Memory | ◦ preprint | arxiv.org/abs/2305.10250 |
| TRIME — Training Language Models with Memory Augmentation | ◦ preprint | arxiv.org/abs/2205.12674 |
| MemAgent — Multi-Conv RL-based Memory Agent for long context | ◦ preprint | arxiv.org/abs/2507.02259 |
| GAM — Hierarchical Graph-based Agentic Memory | ◦ preprint | arxiv.org/abs/2604.12285 |
| Agentic Memory (AgeMem) — Unified Long-/Short-Term Memory Management | ◦ preprint | arxiv.org/abs/2601.01885 |
| GraphRAG — From Local to Global: A Graph RAG Approach to Query-Focused Summarization | — unref | arxiv.org/abs/2404.16130 |
| HippoRAG — Neurobiologically Inspired Long-Term Memory for LLMs | ◦ preprint | arxiv.org/abs/2405.14831 |
| HippoRAG 2 — From RAG to Memory: Non-Parametric Continual Learning for LLMs | ◦ preprint | arxiv.org/abs/2502.14802 |
| LightRAG — Simple and Fast Retrieval-Augmented Generation | ◦ preprint | arxiv.org/abs/2410.05779 |
| RAPTOR — Recursive Abstractive Processing for Tree-Organized Retrieval | ◦ preprint | arxiv.org/abs/2401.18059 |
| Graph Retrieval-Augmented Generation: A Survey | ◦ preprint | arxiv.org/abs/2408.08921 |
| G-Retriever — RAG for Textual Graph Understanding and Question Answering | ◦ preprint | arxiv.org/abs/2402.07630 |
| StructRAG — Boosting Knowledge-Intensive Reasoning via Inference-time Hybrid Information Structurization | ◦ preprint | arxiv.org/abs/2410.08815 |
| NodeRAG — Structuring Graph-based RAG with Heterogeneous Nodes | ◦ preprint | arxiv.org/abs/2504.11544 |
| HeteroRAG — A Heterogeneous RAG Framework for Medical Vision Language Tasks | — unref | arxiv.org/abs/2508.12778 |
| HeteRAG — A Heterogeneous RAG Framework with Decoupled Knowledge Representations | ◦ preprint | arxiv.org/abs/2504.10529 |
| HetaRAG — Hybrid Deep RAG across Heterogeneous Data Stores | ◦ preprint | arxiv.org/abs/2509.21336 |
| Search-R1 — Training LLMs to Reason and Leverage Search Engines with RL | ◦ preprint | arxiv.org/abs/2503.09516 |
| DeepRAG — Thinking to Retrieve Step by Step for LLMs | ◦ preprint | arxiv.org/abs/2502.01142 |
| Search-o1 — Agentic Search-Enhanced Large Reasoning Models | ◦ preprint | arxiv.org/abs/2501.05366 |
| R1-Searcher — Incentivizing the Search Capability in LLMs via RL | ◦ preprint | arxiv.org/abs/2503.05592 |
| ReSearch — Learning to Reason with Search for LLMs via RL | ◦ preprint | arxiv.org/abs/2503.19470 |
| IterDRAG — Inference Scaling for Long-Context Retrieval-Augmented Generation | ◦ preprint | arxiv.org/abs/2410.04343 |
| PlanRAG — Plan-then-Retrieval Augmented Generation for Decision Making | ◦ preprint | arxiv.org/abs/2406.12430 |
| CRAG — Corrective Retrieval Augmented Generation | ◦ preprint | arxiv.org/abs/2401.15884 |
| RankRAG — Unifying Context Ranking with RAG | ◦ preprint | arxiv.org/abs/2407.02485 |
| Speculative RAG — Enhancing RAG through Drafting | ◦ preprint | arxiv.org/abs/2407.08223 |
| Blended RAG — Semantic Search + Hybrid Query-Based Retrievers | ◦ preprint | arxiv.org/abs/2404.07220 |
| Contextual Retrieval — Introducing Contextual Retrieval | anthropic.com/news/contextual-retrieval | |
| Retrieval-Augmented Generation for LLMs: A Survey | ◦ preprint | arxiv.org/abs/2312.10997 |
| Modular RAG — Transforming RAG into LEGO-like Reconfigurable Frameworks | ◦ preprint | arxiv.org/abs/2407.21059 |
| Agentic Retrieval-Augmented Generation: A Survey | ◦ preprint | arxiv.org/abs/2501.09136 |
| Zep — A Temporal Knowledge Graph Architecture for Agent Memory | ◦ preprint | arxiv.org/abs/2501.13956 |
| Larimar — LLMs with Episodic Memory Control | ◦ preprint | arxiv.org/abs/2403.11901 |
| MemInsight — Autonomous Memory Augmentation for LLM Agents | ◦ preprint | arxiv.org/abs/2503.21760 |
| MIRIX — Multi-Agent Memory System for LLM-Based Agents | ◦ preprint | arxiv.org/abs/2507.07957 |
| Generative Agents — Interactive Simulacra of Human Behavior | ◦ preprint | arxiv.org/abs/2304.03442 |
| Sleep-time Compute — Beyond Inference Scaling at Test-time | ◦ preprint | arxiv.org/abs/2504.13171 |
| A Survey on the Memory Mechanism of LLM-based Agents | ◦ preprint | arxiv.org/abs/2404.13501 |
| From Human Memory to AI Memory: A Survey on Memory Mechanisms in the Era of LLMs | ◦ preprint | arxiv.org/abs/2504.15965 |
| From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms | ◦ preprint | arxiv.org/abs/2605.06716 |
| LOCOMO — Evaluating Very Long-Term Conversational Memory of LLM Agents | ◦ preprint | arxiv.org/abs/2402.17753 |
| LongMemEval — Benchmarking Chat Assistants on Long-Term Interactive Memory | ◦ preprint | arxiv.org/abs/2410.10813 |
| CRAG — Comprehensive RAG Benchmark | ◦ preprint | arxiv.org/abs/2406.04744 |
| RAGAS — Automated Evaluation of Retrieval Augmented Generation | ◦ preprint | arxiv.org/abs/2309.15217 |
| ARES — An Automated Evaluation Framework for RAG Systems | ◦ preprint | arxiv.org/abs/2311.09476 |
| RGB — Benchmarking Large Language Models in Retrieval-Augmented Generation | ◦ preprint | arxiv.org/abs/2309.01431 |
| Bridge Evidence: Static Retrieval Utility Does Not Predict Causal Utility in Multi-Step Agentic Search | ◦ preprint | arxiv.org/abs/2607.15253 |
| When Does Continual Learning Require Learning | ◦ preprint | arxiv.org/abs/2607.07847 |
| Beyond Self-Knowledge: Propagating Uncertainty Across Reasoning and Retrieval in LLMs | ◦ preprint | arxiv.org/abs/2607.25600 |
| Eviction as Estimation: A Fixed-Lag Smoothing View of Test-Time Memory, and When Measuring Beats Accumulating | arxiv.org/abs/2607.24667 |