# MemDefrag: Latent Memory Defragmentation for Large Language Models

arXiv: https://arxiv.org/abs/2607.05969 · HTML: https://mlsanity.com/paper/2607.05969/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 41 references that are in the map.
- First posted on arXiv 7 Jul 2026.

## Builds on (most cited first, 23 of 23)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [MemGPT: Towards LLMs as Operating Systems](https://mlsanity.com/paper/2310.08560/) (2023)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [A Survey on the Memory Mechanism of Large Language Model based Agents](https://mlsanity.com/paper/2404.13501/) (2024)
- [Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference](https://mlsanity.com/paper/2406.10774/) (2024)
- [LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression](https://mlsanity.com/paper/2403.12968/) (2024)
- [Layer by Layer: Uncovering Hidden Representations in Language Models](https://mlsanity.com/paper/2502.02013/) (2025)
- [Memory in the Age of AI Agents](https://mlsanity.com/paper/2512.13564/) (2025)
- [Cost-Efficient Large Language Model Serving for Multi-turn Conversations with CachedAttention](https://mlsanity.com/paper/2403.19708/) (2024)
- [Key-value memory in the brain](https://mlsanity.com/paper/2501.02950/) (2025)
- [Pre-training Limited Memory Language Models with Internal and External Knowledge](https://mlsanity.com/paper/2505.15962/) (2025)
- [NextMem: Towards Latent Factual Memory for LLM-based Agents](https://mlsanity.com/paper/2603.15634/) (2026)
- [RePo: Language Models with Context Re-Positioning](https://mlsanity.com/paper/2512.14391/) (2025)
- [Time is Not a Label: Continuous Phase Rotation for Temporal Knowledge Graphs and Agentic Memory](https://mlsanity.com/paper/2604.11544/) (2026)
- [SeqPE: Transformer with Sequential Position Encoding](https://mlsanity.com/paper/2506.13277/) (2025)
- [Semantic Chunking and the Entropy of Natural Language](https://mlsanity.com/paper/2602.13194/) (2026)

## Abstract (arXiv)

Latent memory, which stores past knowledge fragments as per-layer hidden states, has emerged as a promising paradigm (e.g., MemoryLLM and M+) for long-term memory in large language models (LLMs). However, the paradigm suffers from significant performance degradation during memory updates, due to positional encoding misalignment and the absence of any tracing mechanism to distinguish target memory fragments from irrelevant ones. To discover such a tracing mechanism, we probe the layer-wise attention density over stored memory fragments, and find that a small set of middle transformer layers consistently concentrates the highest density on the target fragment - exposing an inherent tracing signal. In light of this, we propose MemDefrag, a training-free and model-agnostic framework that (1) uses a middle-layer tracing signal to conduct memory defragmentation (rank, reorder, and filter memories), and (2) applies an informativeness-guided proportional forgetting mechanism once capacity is exceeded. Experiments show that MemDefrag substantially outperforms MemoryLLM and M+ on knowledge retention (e.g., 43.0% vs. 17.4%/17.6% after 50 memory updates) and long-context benchmarks, and generalizes well across various LLMs and latent-memory variants. The code is available at github.com/ryehr/MemDefrag.