{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WECPCQBUSPNG5JFT5XFRFZKSRV","short_pith_number":"pith:WECPCQBU","schema_version":"1.0","canonical_sha256":"b104f1403493da6ea4b3edcb12e5528d469b423c47b7eed6f1baff2b7abc58cf","source":{"kind":"arxiv","id":"2505.22101","version":1},"attestation_state":"computed","paper":{"title":"MemOS: An Operating System for Memory-Augmented Generation (MAG) in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bo Tang, Chenyang Xi, Ding Chen, Feiyu Xiong, Hanyu Wang, Hongkang Yang, Huayi Lai, Jiahao Huo, Jiawei Yang, Jihao Zhao, Junpeng Ren, Kai Chen, Kehang Li, Qingchen Yu, Shichao Song, Simin Niu, Tianyi Chen, Yezhaohui Wang, Zehao Lin, Zhiqiang Yin, Zhi-Qin John Xu, Zhiyu Li","submitted_at":"2025-05-28T08:27:12Z","abstract_excerpt":"Large Language Models (LLMs) have emerged as foundational infrastructure in the pursuit of Artificial General Intelligence (AGI). Despite their remarkable capabilities in language perception and generation, current LLMs fundamentally lack a unified and structured architecture for handling memory. They primarily rely on parametric memory (knowledge encoded in model weights) and ephemeral activation memory (context-limited runtime states). While emerging methods like Retrieval-Augmented Generation (RAG) incorporate plaintext memory, they lack lifecycle management and multi-modal integration, lim"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2505.22101","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-28T08:27:12Z","cross_cats_sorted":[],"title_canon_sha256":"8bb33fdf7f0204b816410ab724d96b42d397c1e480f54997006fe17693228d09","abstract_canon_sha256":"d5d68fcc3fb4ccbfca366bf425bbd4f11cae33356055e5b47270c7b629315af4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:09.926466Z","signature_b64":"AkgXbe5Q8pzZGml/ocVWBIMUIZ8eztnCgkdlBLIdpfLoyoMiOtzHurB1PXXM2xMI4nMHGRRSuSiZWrZxyw7IBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b104f1403493da6ea4b3edcb12e5528d469b423c47b7eed6f1baff2b7abc58cf","last_reissued_at":"2026-07-05T11:11:09.925888Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:09.925888Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MemOS: An Operating System for Memory-Augmented Generation (MAG) in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bo Tang, Chenyang Xi, Ding Chen, Feiyu Xiong, Hanyu Wang, Hongkang Yang, Huayi Lai, Jiahao Huo, Jiawei Yang, Jihao Zhao, Junpeng Ren, Kai Chen, Kehang Li, Qingchen Yu, Shichao Song, Simin Niu, Tianyi Chen, Yezhaohui Wang, Zehao Lin, Zhiqiang Yin, Zhi-Qin John Xu, Zhiyu Li","submitted_at":"2025-05-28T08:27:12Z","abstract_excerpt":"Large Language Models (LLMs) have emerged as foundational infrastructure in the pursuit of Artificial General Intelligence (AGI). Despite their remarkable capabilities in language perception and generation, current LLMs fundamentally lack a unified and structured architecture for handling memory. They primarily rely on parametric memory (knowledge encoded in model weights) and ephemeral activation memory (context-limited runtime states). While emerging methods like Retrieval-Augmented Generation (RAG) incorporate plaintext memory, they lack lifecycle management and multi-modal integration, lim"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.22101","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.22101/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2505.22101","created_at":"2026-07-05T11:11:09.925971+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.22101v1","created_at":"2026-07-05T11:11:09.925971+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.22101","created_at":"2026-07-05T11:11:09.925971+00:00"},{"alias_kind":"pith_short_12","alias_value":"WECPCQBUSPNG","created_at":"2026-07-05T11:11:09.925971+00:00"},{"alias_kind":"pith_short_16","alias_value":"WECPCQBUSPNG5JFT","created_at":"2026-07-05T11:11:09.925971+00:00"},{"alias_kind":"pith_short_8","alias_value":"WECPCQBU","created_at":"2026-07-05T11:11:09.925971+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":31,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05794","citing_title":"From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2606.22385","citing_title":"MetaPS: Adaptive Programmatic Strategy Selection for Market Agents","ref_index":137,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19172","citing_title":"User as Engram: Internalizing Per-User Memory as Local Parametric Edits","ref_index":63,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10662","citing_title":"Decentralized Multi-Agent Systems with Shared Context","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10949","citing_title":"Recalling Too Well: Sycophancy Evaluation and Mitigation in Memory-Augmented Models","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09365","citing_title":"Experience Makes Skillful: Enabling Generalizable Medical Agent Reasoning via Self-Evolving Skill Memory","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07711","citing_title":"Rosetta Memory: Adaptive Memory for Cross-LLM Agents","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05761","citing_title":"SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04780","citing_title":"PersonaTree: Structured Lifecycle Memory for Person Understanding in LLM Agents","ref_index":72,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04815","citing_title":"Learning While Acting: A Skill-Enhanced Test-Time Co-Evolution Framework for Online Lifelong Learning Agents","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14906","citing_title":"MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29788","citing_title":"MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25002","citing_title":"MemMark: State-Evolution Attribution Watermarking for Agent Long-Term Memory Systems","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25379","citing_title":"EfficientGraph-RAG: Structured Retrieval-State Management for Cross-Task Retrieval-Augmented Generation","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28009","citing_title":"MemGuard: Preventing Memory Contamination in Long-Term Memory-Augmented Large Language Models","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29640","citing_title":"VikingMem: A Memory Base Management System for Stateful LLM-based Applications","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30159","citing_title":"Meta-Cognitive Memory Policy Optimization for Long-Horizon LLM Agents","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30260","citing_title":"How LoRA Remembers? A Parametric Memory Law for LLM Finetuning","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11759","citing_title":"Retrieval Is Not Enough: Why Organizational AI Needs Epistemic Infrastructure","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21768","citing_title":"Memory-R2: Fair Credit Assignment for Long-Horizon Memory-Augmented LLM Agents","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17065","citing_title":"PyraVid: Hierarchical Multimodal Memory for Long-Horizon Video Reasoning","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01970","citing_title":"Trojan Hippo: Weaponizing Agent Memory for Data Exfiltration","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15384","citing_title":"Is One Score Enough? Rethinking the Evaluation of Sequentially Evolving LLM Memory","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2601.12538","citing_title":"Agentic Reasoning for Large Language Models","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02522","citing_title":"Opal: Private Memory for Personal AI","ref_index":129,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WECPCQBUSPNG5JFT5XFRFZKSRV","json":"https://pith.science/pith/WECPCQBUSPNG5JFT5XFRFZKSRV.json","graph_json":"https://pith.science/api/pith-number/WECPCQBUSPNG5JFT5XFRFZKSRV/graph.json","events_json":"https://pith.science/api/pith-number/WECPCQBUSPNG5JFT5XFRFZKSRV/events.json","paper":"https://pith.science/paper/WECPCQBU"},"agent_actions":{"view_html":"https://pith.science/pith/WECPCQBUSPNG5JFT5XFRFZKSRV","download_json":"https://pith.science/pith/WECPCQBUSPNG5JFT5XFRFZKSRV.json","view_paper":"https://pith.science/paper/WECPCQBU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.22101&json=true","fetch_graph":"https://pith.science/api/pith-number/WECPCQBUSPNG5JFT5XFRFZKSRV/graph.json","fetch_events":"https://pith.science/api/pith-number/WECPCQBUSPNG5JFT5XFRFZKSRV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WECPCQBUSPNG5JFT5XFRFZKSRV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WECPCQBUSPNG5JFT5XFRFZKSRV/action/storage_attestation","attest_author":"https://pith.science/pith/WECPCQBUSPNG5JFT5XFRFZKSRV/action/author_attestation","sign_citation":"https://pith.science/pith/WECPCQBUSPNG5JFT5XFRFZKSRV/action/citation_signature","submit_replication":"https://pith.science/pith/WECPCQBUSPNG5JFT5XFRFZKSRV/action/replication_record"}},"created_at":"2026-07-05T11:11:09.925971+00:00","updated_at":"2026-07-05T11:11:09.925971+00:00"}