LiveMem: Maintaining Memory State Continuity in Long-Running LLM Inference
- URL: http://arxiv.org/abs/2608.02515v2
- Date: Fri, 07 Aug 2026 09:41:07 GMT
- Title: LiveMem: Maintaining Memory State Continuity in Long-Running LLM Inference
- Abstract summary: Long-running assistants and agents consume computation streams that eventually outgrow the context.<n>Existing context retention, summarization, and retrieval preserve access to selected history, but do not provide a persistent state over the full lifecycle when working context changes.<n>We introduce an intrinsic memory method, textbfLiveMem, which augments a pretrained full-attention LLM with a memory state that preserves the historical information over the whole lifecycle.
- Score: 6.324239367257621
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Long-running assistants and agents consume interaction streams that eventually outgrow the context. Existing context retention, summarization, and retrieval preserve access to selected history, but do not provide a persistent state over the full lifecycle when working context changes. We formulate this missing inference capability as \emph{state continuity under context turnover}: carrying computation forward through a fixed-capacity memory state whose lifetime is independent of the active context. We introduce an intrinsic memory method, \textbf{LiveMem}, which augments a pretrained full-attention LLM with a memory state that preserves the historical information over the whole lifecycle while the main attention path retains a bounded KV window. Context turnover and memory state maintaining, memory-oriented post-training, and state-aware serving jointly make this memory state load bearing after its originating tokens are released. Our experiments show that LiveMem achieves leading overall performance among evaluated systems and other intrinsic memory methods. Experiments on LongMemEval show that LiveMem is able to answer the question based on the memory state, even when the supporting evidence has been removed from the current context, and evidence-distance analysis shows that useful information persists beyond the active window. LiveMem thus establishes state continuity as a distinct and complementary abstraction for continual LLM inference.
Related papers
- LifeFuse-Mem: Lifecycle-Aware State Fusion Against Temporary Overwriting for Long-Term Memory [16.775600156252096]
LLM agents require memory mechanisms that maintain coherent internal states across interactions.<n>LifeFuse-Mem is a neural memory framework that separates information according to its temporal commitment.<n>LifeFuse-Mem improves acquisition-controlled retention and reduces temporary overwrite.
arXiv Detail & Related papers (2026-09-11T04:56:48Z) - Beyond Retrieval: Progressive Latent Memory Evolution for Streaming Video Understanding [64.34948094316685]
We introduce LatentStream, a progressive latent working memory framework that shifts streaming memory from store-and-retrieve to retrieve-and-internalize.<n>LatentStream achieves new state-of-the-art results on existing online and offline video benchmarks.
arXiv Detail & Related papers (2026-09-03T17:28:14Z) - MemPrism: Task-Conditioned Relational Memory Views for Long-Horizon Agents [14.296334759650918]
Long-horizon agents rely on memory to reuse experiences, yet existing memory systems assume that evidence can be directly consumed through a fixed representation.<n>MemPrism is a task-conditioned relational memory framework that separates persistent experience storage from decision-time working memory.<n>MemPrism consistently improves the task performance, especially as trajectories become longer, while reducing memory token consumption.
arXiv Detail & Related papers (2026-08-07T03:13:43Z) - UniMem: Complementary Episodic-to-Parametric Memory for Boundary-Agnostic Task Streams [77.52140119830241]
We propose UniMem, a self-routing framework for autonomous memory management.<n>UniMem uses learnable routing tokens as memory controllers, enabling adaptive coordination between complementary memory pathways.<n>Experiments on long-horizon streaming task sequences show that UniMem consistently outperforms baselines while maintaining execution fidelity.
arXiv Detail & Related papers (2026-07-28T17:28:21Z) - Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents [72.3752981234916]
Large Language Model (LLM) agents increasingly rely on external memory systems to accumulate experience across tasks.<n>We present Memory as a Controlled Process (MemCon), a framework that models memory operations as a Markov Decision Process.<n>MemCon consistently outperforms multiple memory baselines by up to 15.2 points in task success while reducing token consumption by 5-20%.
arXiv Detail & Related papers (2026-07-15T08:32:40Z) - MemOps: Benchmarking Lifecycle Memory Operations in Long-Horizon Conversations [50.24315431387575]
We introduce MemOps, a benchmark that reformulates conversational memory as a sequence of lifecycle operations.<n>MemOps represents each memory event with a structured trace specifying its trigger, target, scope, state transition, and supporting evidence.<n>Across long-context, retrieval-based, parametric and managed-memory systems, MemOps disentangles failure modes that final-answer accuracy alone conceals.
arXiv Detail & Related papers (2026-07-14T15:33:44Z) - Is One Score Enough? Rethinking the Evaluation of Sequentially Evolving LLM Memory [50.857546269660276]
We introduce SeqMem-Eval, a diagnostic evaluation framework for sequentially evolving large language models (LLMs) memory.<n>Rather than focusing only on final performance, SeqMem-Eval evaluates how memory states evolve, generalize, consolidate experience, and retain useful information during sequential inference.
arXiv Detail & Related papers (2026-05-14T20:15:22Z) - Mem-W: Latent Memory-Native GUI Agents [50.87647372904382]
We introduce Mem-W, a series of latent-memory-native GUI agents that treat memory as part of the agent's continuous context.<n>Across four web and mobile navigation benchmarks, Mem-W consistently improves diverse backbones and memory-enhanced baselines.
arXiv Detail & Related papers (2026-05-10T04:31:23Z) - Learning to Remember: End-to-End Training of Memory Agents for Long-Context Reasoning [18.621823772319154]
We propose an end-to-end reinforcement learning framework that unifies memory operations and question answering within a single policy.<n>UMA maintains a dual memory representation: a compact core summary for global context and a structured Memory Bank that supports explicit CRUD.<n>Across 13 datasets spanning Ledger-QA, Test-Time Learning, and Accurate Retrieval, UMA substantially outperforms long-context and RAG baselines on dynamic reasoning and learning tasks.
arXiv Detail & Related papers (2026-02-13T16:54:23Z) - EverMemOS: A Self-Organizing Memory Operating System for Structured Long-Horizon Reasoning [42.339841548168565]
Large Language Models (LLMs) are increasingly deployed as long-term interactive agents, yet their limited context windows make it difficult to sustain coherent behavior over extended interactions.<n>We introduce EverMemOS, a self-organizing memory operating system that implements an engram-inspired lifecycle for computational memory.<n>EverMemOS achieves state-of-the-art performance on memory-augmented reasoning tasks.
arXiv Detail & Related papers (2026-01-05T14:39:43Z) - Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory [89.65731902036669]
Evo-Memory is a streaming benchmark and framework for evaluating self-evolving memory in large language model (LLM) agents.<n>We evaluate over ten representative memory modules and evaluate them across 10 diverse multi-turn goal-oriented and single-turn reasoning and QA datasets.
arXiv Detail & Related papers (2025-11-25T21:08:07Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.