MemLife: Curating and Reasoning over Long-Term Egocentric Video Memories
Summary
MemLife introduces a multimodal memory system for long-term egocentric videos that builds entity-grounded text episodes and retrieves them with a time-indexed agentic reader, improving over training-free baselines by 4.6-12.0% on long-horizon benchmarks. A reinforcement learning framework called MemOpt further optimizes the memory writer for faithfulness and retrievability, yielding consistent 2.7-5.0% gains.
View Cached Full Text
Cached at: 10/01/26, 04:23 PM
Paper page - MemLife: Curating and Reasoning over Long-Term Egocentric Video Memories
Source: https://huggingface.co/papers/2609.40195 Authors:
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
Abstract
Long-termegocentricvideoenablespersonalizedAIassistantstoreasonaboutdailylife.However,asvideohistoriesgrowtohundredsofhoursspanningmonthsoryears,reprocessingrawclipsforeveryquerybecomescomputationallyprohibitive.Memorysystemsofferascalablealternativebycompactingvideosintotextrepresentations,butoftenfailonpracticalbenchmarks:eitherthememorydoesnotpreservekeyevidence,ortheretrieverfailstolocaterelevantentriesduetoretrievalcompetitioningrowingsearchspaces.Toaddressthesechallenges,weintroduceMemLife,amultimodalmemorysystemthatconstructsentity-grounded,first-persontextepisodesandretrievesthemviaatime-indexedagenticreader.Withouttrainingorquery-timevideoaccess,MemLifeimprovesoverthestrongesttraining-freebaselineby4.6--12.0%acrossfourlong-horizonbenchmarks.Tofurtherimprovememoryquality,weproposeMemOpt,areinforcementlearningframeworkthatoptimizesthememorywritertoproducefaithful,informative,andretrievablememories.MemOptconsistentlyimprovesMemLifeby2.7--5.0%acrossdifferentvideoandquestiondistributions,withgainsthatgeneralizeacrosswriterandreaderbackbonesandmemorysystems.
View arXiv pageView PDFAdd to collection
Get this paper in your agent:
hf papers read 2609\.40195
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2609.40195 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2609.40195 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2609.40195 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
MemDreamer: Decoupling Perception and Reasoning for Long Video Understanding via Hierarchical Graph Memory and Agentic Retrieval Mechanism
MemDreamer decouples perception and reasoning for long video understanding using hierarchical graph memory and agentic retrieval, achieving state-of-the-art performance with reduced computational overhead.
MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory
MemEye is a visual-centric evaluation framework that assesses multimodal agent memory by measuring visual evidence granularity and retrieval complexity across 8 life-scenario tasks, revealing that current architectures struggle to preserve fine-grained visual details and reason about state changes over time.
MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models
MemLens is a new benchmark for evaluating memory capabilities in large vision-language models through multi-session conversations. It compares long-context and memory-augmented approaches, revealing limitations in both and motivating hybrid architectures.
MemFit: Efficient Long-Term Agentic Memory
MemFit is a long-term memory system for conversational agents that stores turns verbatim in an append-only store and uses LLM-free insertion and multi-path retrieval to drastically cut memory construction cost and latency while achieving state-of-the-art results on LoCoMo, MemGallery, and LongMemEval-S.
Reasoning with Memory: A Temporal Granularity-Adaptive Framework for Training-Free Long Video Understanding
ReMem introduces a dual-level memory-augmented keyframe selection framework for training-free long video understanding, achieving state-of-the-art zero-shot performance on multiple benchmarks.