Memo

Follow this topic Ask about this topic

Momentum

5 papers in the last four weeks, up 67% on the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 27

All topics

Memorizon: Training World Models Beyond Their Context Window
Sep 30, 2026Tingting Liao, Xuezhi Liang, Hao Li +1Memory-Augmented Framework Long-Term Memory
IFM, MBZUAI · MBZUAI
MEMO: Multi-Level Entity-Aware Memory for Streaming Video Understanding
Sep 30, 2026Yinying Li, Yuqian Fu, Yulin Dai +3Streaming Video Understanding Streaming Video
East China Normal University · King Abdullah University of Science and Technology
Single-Layer MeMo as a Randomized Hamming-Kernel Classifier
Sep 28, 2026Alessandro StraziotaMultilayer Perceptrons Language Modeling
Department of Enterprise Engineering University of Rome “Tor Vergata” Rome, Italy
EdgeReMIND: A Scalable, Top-Ranked Memorization Baseline for Temporal Multi-Relational Link Prediction
Sep 15, 2026Bryant PollardZero-Shot Link Prediction Memo
Clemson University
MEMO: Multimodal Evidence Memory Organization for Long-Horizon LLM Agents
Sep 7, 2026Xian Gao, Jinpeng Wang, Jiacheng Ruan +3Multimodal Memory Memo
Shanghai Jiao Tong University · Tsinghua University · Institute of Automation of the Chinese Academy of Sciences
RippleMem: From Isolated Retrieval to Associative Recollection for Long-Term Agent Memory
Aug 13, 2026Jingbo Ji, Lingyi Li, Xilong Cheng +4Long-Term Agent Memory Long-Term Memory
Communication University of China, Beijing, China. · Zhilian Yinghe Technology Co., Ltd., Beijing, China · State Key Laboratory of Media Convergence and Communication, Beijing, China
MindMemOS: A Portable and Self-Evolving Memory Operating Layer for AI Agents
Aug 12, 2026Kaichao Liang, Yuqi Cui, Hao Kong +13Self-Evolving Agents Memo
Noah’s Ark Lab, Huawei Technologies
MemeMind: Reference-Guided Trace Construction for Offline Context Optimization
Aug 10, 2026Run Yang, Weihang Wang, Boheng Sheng +7Memo Contextual Reasoning
1Bilibili · 2Fudan University
ConMem: Contribution-Aware Memory for Long-Horizon Manufacturing Inspection Logs
Jul 30, 2026Bingchen Liu, Yuanyuan Fang, Lei Liu +6Memo Inspection
Shandong University, School of Software Jinan, Shandong, China · Boston University, Metropolitan College Boston, MA, USA · North China Electric Power University, School of Control and Computer Engineering Beijing, Beijing, China +3
MEMENTO: Memory-Guided Memetic Code-as-Policy Evolution
Jul 24, 2026Alkis Sygkounas, Victor Aregbede, Amy Loutfi +1On-Policy Self-Evolution Robot Policies
Center for Applied Autonomous Sensor Systems (AASS), Örebro University, Sweden · Örebro University, 70281 Örebro, Sweden · Linköing University, 58183 Linköping, Sweden
VisionAssist: An Open-Source Smartphone Assistant for AI-Based Visual Accessibility
Jul 18, 2026Ayşe Özlem Çalışkan, Jordi Sanchez-RieraAssistive Agents Accessibility
Institut de Robòtica i Informàtica Industrial (CSIC-UPC)
MEMORA: Embodied Action Memory from Egocentric Videos for Reasoning and Planning
Jul 15, 2026Zihao Yu, Xiu Yuan, Chongjie ZhangClassical Planning Memo
Washington University in St. Louis
REMIND: RE-Identification with Memory for INDoor Navigation
Jul 10, 2026Pablo Diaz-Pereda, Alejandro Rodriguez-Ramos, David Perez-Saura +1Multi-Object Tracking Re-Identification
Computer Vision and Aerial Robotics Group at Centre for Automation and Robotics C.A.R. (UPM-CSIC), Universidad Politécnica de Madrid (CVAR-UPM), Calle Jose Gutierrez Abascal 2, 28006 Madrid, Spain.
ComMem: Complementary Memory Systems for Test-Time Adaptation of Vision-Language Models
Jun 27, 2026Guanglong Sun, Shuang Cui, Bo Lei +6Vision-Language Model Adaptation Test-Time Adaptation
School of Life Sciences, IDG/McGovern Institute for Brain Research, Tsinghua University, Beijing, China · Institute of Software Chinese Academy of Sciences, Beijing, China · 3Beijing Academy of Artificial Intelligence, Beijing, China +2
Staying In Character: Perspective-Bounded Memory For Book-Based Role-Playing Agents
Jun 24, 2026Xushuo Tang, Junhe Zhang, Zihan Yang +4Role-Playing Agents Narrative Generation
UNSW Sydney · Chang’an University · RAIDS Lab +2
Towards Version-aware Operations and Transaction Memories for Multi-layer MeMo
Jun 23, 2026Peiran LiVersion Memo
Freie Universität Berlin, Kaiserswerther Str. 16-18, 14195 Berlin, Germany
Memento: Reconstruct to Remember for Consistent Long Video Generation
Jun 12, 2026Xuan Wei, Longbin Ji, Guan Wang +5Visual Memory Long-Term Memory
Xiamen University · ERNIE Team, Baidu Inc.
ConMem: Structured Memory-Guided Adaptation in Training-Free Multi-Agent Systems
Jun 7, 2026Zhixun Tan, Qiang Chen, Tairan Huang +2Graph-Structured Memories Memo
Central South University · The Hong Kong University of Science and Technology
MEMENTO: Leveraging Web as a Learning Signal for Low-Data Domains
May 28, 2026Ashutosh Ojha, Vinay Aggarwal, Ashutosh Srivastava +3Web Few-Shot Learning
Adobe, Media & Data Science Research Lab
FundaPod: A Multi-Persona Agent Pod Architecture with Knowledge Graph Memory for AI-Assisted Fundamental Investment Research
May 27, 2026Di Zhu, Lei Nico Zheng, Zihan ChenFinancial Strategy Research Memo
Stevens Institute of Technology Hoboken, NJ, USA · UMass Boston Boston, MA, USA
RecMem: Recurrence-based Memory Consolidation for Efficient and Effective Long-Running LLM Agents
May 15, 2026Zijie Dai, Shiyuan Deng, Sheng Guan +4Semantic Memory Memo
Department of Computer Science and Engineering, The Chinese University of Hong Kong · Huawei Cloud · School of Computer Science, Beijing University of Posts and Telecommunications +2
Memory Inception: Latent-Space KV Cache Manipulation for Steering LLMs
May 7, 2026Andy Zeyi Liu, Michael Zhang, Ilana Greenberg +3Linear Activation Steering Large Language Model Memory
Yale University · Princeton University · Jump Trading
Contextual Agentic Memory is a Memo, Not True Memory
Apr 30, 2026Binyan Xu, Xilin Dai, Kehuan ZhangAgentic Memory Memo
The Chinese University of Hong Kong, Hong Kong, China · Zhejiang University, Hangzhou, China
MEMOR-E: In-Context and Fine-Tuned LLM Personalization for Alzheimer's Assistive Robotics
Apr 28, 2026Maissa Abir Smaili, Eren Sadikoglu, Ransalu SenanayakeAlzheimer Cognitive Diagnosis
Istanbul Medipol University Istanbul, Türkiye · Arizona State University Tempe, USA
From Recall to Forgetting: Benchmarking Long-Term Memory for Personalized Agents
Apr 21, 2026Md Nayem Uddin, Kumar Shubham, Eduardo Blanco +2Long-Term Agent Memory Agentic Memory
Arizona State University · Genies · University of Arizona
Memora: A Harmonic Memory Representation Balancing Abstraction and Specificity
Feb 3, 2026Menglin Xia, Xuchao Zhang, Shantanu Dixit +6Memo
M365 Research, Microsoft.
Memento No More: Coaching AI Agents to Master Multiple Tasks via Hints Internalization
Feb 3, 2025Minttu Alakuijala, Ya Gao, Georgy Ananov +4Artificial Intelligence Agents Internalization
Department of Computer Science, Aalto University · Department of Computer Science, University of Manchester · System 2 AI

Sep 30, 2026, cs.CVJ/K move · Enter open · S save

Memorizon: Training World Models Beyond Their Context Window

Tingting Liao, Xuezhi Liang, Hao Li, Guangyi Liu

IFM, MBZUAI · MBZUAI

Memory-Augmented Framework Long-Term Memory Memo Model Training Streaming

Open paper AskPDF arXiv

Streaming world models should render a place consistently across repeated visits. Directly supervising such revisits requires training samples that capture both visits, often spanning minutes. Yet dense attention over the full span incurs quadratic costs, making long-span supervision expensive. Memorizon breaks this coupling: long spans are needed for supervision, but not for attention, since the two visits can share a forward pass without including every intervening frame. A training sample covers a span of any length but is scored only on its last

k

chunks. Instead of tokenizing the history before them, each scored chunk retrieves its own top-

K

latents by camera co-visibility, and the union of these requests forms a shared bank. The bank is bounded by

kK

, so the sequence stays bounded however long the span; at the shortest span the recipe is exactly conventional training. Adding the bank raises the cost of a step once; beyond that, a longer span costs little, and going from 100 to 400 s adds 12% to the step time. Against a sliding-window baseline, retrieval raises revisit consistency on every split, and a span long enough to reach the first visit of each return adds a further 24% to 30%, at some cost in image quality; beyond that span, more length no longer helps. Filling the bank from another episode lowers revisit correlation by 83%, so the model uses what it retrieves. Project page: https://tingtingliao.github.io/memorizon

Momentum

Related topics

Latest papers 27