microsoft/KBLaM

The official implementation of the paper "MLP Memory: A Retriever-Pretrained Memory for Large Language Models". (ICLR 2026)

Cache-Augmented Generation: A Simple, Efficient Alternative to RAG

Data and Code for EMNLP 2025 Findings Paper "MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search"

The official repo of the paper: $delta$-mem: Efficient Online Memory for Large Language Models

This includes the original implementation of SELF-RAG: Learning to Retrieve, Generate and Critique through self-reflection by Akari Asai, Zeqiu Wu, Yizhong Wang, Avirup Sil, and Hannaneh Hajishirzi.

Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models

Keeping language models honest by directly eliciting knowledge encoded in their activations.

(AAAI 2026) Knowledge-Graph-Based Generation with Semantic Aggregation and Hierarchical Retrieval