Pith. sign in

REVIEW 10 cited by

Cognitive Memory in Large Language Models

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.02441 v2 pith:5RRD27MG submitted 2025-04-03 cs.CL cs.AI

Lianlei Shan , Shixian Luo , Zezhou Zhu , Yu Yuan , Yong Wu This is my paper · ORCID
classification cs.CLcs.AI
keywords memorymechanismscompressionmethodsapproachesefficiencylanguagelarge
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

This paper examines memory mechanisms in Large Language Models (LLMs), emphasizing their importance for context-rich responses, reduced hallucinations, and improved efficiency. It categorizes memory into sensory, short-term, and long-term, with sensory memory corresponding to input prompts, short-term memory processing immediate context, and long-term memory implemented via external databases or structures. The text-based memory section covers acquisition (selection and summarization), management (updating, accessing, storing, and resolving conflicts), and utilization (full-text search, SQL queries, semantic search). The KV cache-based memory section discusses selection methods (regularity-based summarization, score-based approaches, special token embeddings) and compression techniques (low-rank compression, KV merging, multimodal compression), along with management strategies like offloading and shared attention mechanisms. Parameter-based memory methods (LoRA, TTT, MoE) transform memories into model parameters to enhance efficiency, while hidden-state-based memory approaches (chunk mechanisms, recurrent transformers, Mamba model) improve long-text processing by combining RNN hidden states with current methods. Overall, the paper offers a comprehensive analysis of LLM memory mechanisms, highlighting their significance and future research directions.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 10 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. StreamFlow: Dynamic Memory Flows for Streaming Video Understanding

    cs.CV 2026-08 conditional novelty 6.0 of 10

    A streaming-video memory system that filters redundant frames before encoding, stores older video as latent tokens, and re-injects them when visual attention drops reaches 67.73% on StreamingBench.

  2. ECHO: Prune To Act, Trace To Learn With Selective Turn Memory In Agentic RL

    cs.LG 2026-06 unverdicted novelty 6.0 of 10

    Source-indexed selective turn memory plus provenance-guided credit assignment raises BrowseComp-Plus accuracy to 43.4% over GRPO and SUPO while cutting turn proliferation.

  3. The Future is Agentic: Definitions, Perspectives, and Open Challenges of Multi-Agent Recommender Systems

    cs.IR 2025-07 conditional novelty 6.0 of 10

    A framework for agentic recommender systems plus a pilot study showing multi-agent pipelines beat a single-shot LLM only on high-diversity user histories.

  4. How Can Input Reformulation Improve Tool Usage Accuracy in a Complex Dynamic Environment? A Study on $\tau$-bench

    cs.CL 2025-08 conditional novelty 5.0 of 10

    IRMA reformulates tool-agent inputs with memory, domain constraints, and tool suggestions, and reports improved tau-bench pass^5 reliability over ReAct, function calling, and self-reflection.

  5. MemTool: Optimizing Short-Term Memory Management for Dynamic Tool Calling in LLM Agent Multi-Turn Conversations

    cs.CL 2025-07 conditional novelty 5.0 of 10

    MemTool is a short-term memory framework with three modes (autonomous, workflow, hybrid) that lets LLM agents add and remove tools across multi-turn conversations, evaluated over 100 turns on 13+ models.

  6. GDGS: 3D Gaussian Splatting Via Geometry-Guided Initialization And Dynamic Density Control

    cs.CV 2025-07 conditional novelty 4.0 of 10

    A 3DGS variant that adds MLP initialization, normal alignment, and region-aware density control reports consistent but modest quality gains over vanilla 3DGS on three standard benchmarks.

  7. Building Lightweight Semantic Segmentation Models for Aerial Images Using Dual Relation Distillation

    eess.IV 2025-06 conditional novelty 4.0 of 10

    A student segmentation network trained with spatial and channel relation distillation from a PSPNet ResNet101 teacher gains about 3 to 5 mIoU points on Vaihingen, Potsdam, and Cityscapes.

  8. MemOS: An Operating System for Memory-Augmented Generation (MAG) in Large Language Models

    cs.CL 2025-05 reject novelty 4.0 of 10

    A unified memory-operating-system design for LLMs, built around a MemCube abstraction, is presented without any experimental validation.

  9. Memory-Augmented Transformers: A Systematic Review from Neuroscience Principles to Enhanced Model Architectures

    cs.LG 2025-08 unverdicted novelty 3.0 of 10

    Memory-augmented Transformer research is organized into a three-axis taxonomy bridging neuroscience memory concepts to network designs, but no new result is produced.

  10. A Global-Local Cross-Attention Network for Ultra-high Resolution Remote Sensing Image Semantic Segmentation

    cs.CV 2025-06 reject novelty 3.0 of 10

    GLCANet is a dual-branch global-local attention network that reports top mIoU on DeepGlobe, Vaihingen, and Potsdam, but the method and experiments are internally inconsistent and lack code.

Pith tools