{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:4FWAI53MOHIDRGTQQR7VJTUV52","short_pith_number":"pith:4FWAI53M","schema_version":"1.0","canonical_sha256":"e16c04776c71d0389a70847f54ce95eeb68d316255f1b6d4bca81524852d3206","source":{"kind":"arxiv","id":"2006.11527","version":2},"attestation_state":"computed","paper":{"title":"Memory Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NE"],"primary_cat":"cs.CL","authors_text":"Anton Peganov, Grigory V. Sapunov, Mikhail S. Burtsev, Yuri Kuratov","submitted_at":"2020-06-20T09:06:27Z","abstract_excerpt":"Transformer-based models have achieved state-of-the-art results in many natural language processing tasks. The self-attention architecture allows transformer to combine information from all elements of a sequence into context-aware representations. However, information about the context is stored mostly in the same element-wise representations. This might limit the processing of properties related to the sequence as a whole more difficult. Adding trainable memory to selectively store local as well as global representations of a sequence is a promising direction to improve the Transformer model"},"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":"2006.11527","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-06-20T09:06:27Z","cross_cats_sorted":["cs.LG","cs.NE"],"title_canon_sha256":"73ef9823e6573f1eb5fb64d638d707e0985e57cf9aec5ab6a8e82cafc351fe3e","abstract_canon_sha256":"71413ab4c65a2c98b06bdc8193fc69869470707e3ca184c4f5e19ee36a38a1f6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:15:25.159711Z","signature_b64":"59WbhvYwoyPPBq5kW6O8iyf1shkrZveA2Ni8WjhgZ7UwQEtcJmeU846FI7RRFaxdwnDoh39GIGbsVoZbCONVCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e16c04776c71d0389a70847f54ce95eeb68d316255f1b6d4bca81524852d3206","last_reissued_at":"2026-07-05T02:15:25.159289Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:15:25.159289Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Memory Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NE"],"primary_cat":"cs.CL","authors_text":"Anton Peganov, Grigory V. Sapunov, Mikhail S. Burtsev, Yuri Kuratov","submitted_at":"2020-06-20T09:06:27Z","abstract_excerpt":"Transformer-based models have achieved state-of-the-art results in many natural language processing tasks. The self-attention architecture allows transformer to combine information from all elements of a sequence into context-aware representations. However, information about the context is stored mostly in the same element-wise representations. This might limit the processing of properties related to the sequence as a whole more difficult. Adding trainable memory to selectively store local as well as global representations of a sequence is a promising direction to improve the Transformer model"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.11527","kind":"arxiv","version":2},"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/2006.11527/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":"2006.11527","created_at":"2026-07-05T02:15:25.159354+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.11527v2","created_at":"2026-07-05T02:15:25.159354+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.11527","created_at":"2026-07-05T02:15:25.159354+00:00"},{"alias_kind":"pith_short_12","alias_value":"4FWAI53MOHID","created_at":"2026-07-05T02:15:25.159354+00:00"},{"alias_kind":"pith_short_16","alias_value":"4FWAI53MOHIDRGTQ","created_at":"2026-07-05T02:15:25.159354+00:00"},{"alias_kind":"pith_short_8","alias_value":"4FWAI53M","created_at":"2026-07-05T02:15:25.159354+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.26745","citing_title":"Deep sequence models tend to memorize geometrically; it is unclear why","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2412.13171","citing_title":"Compressed Chain of Thought: Efficient Reasoning Through Dense Representations","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2507.07957","citing_title":"MIRIX: Multi-Agent Memory System for LLM-Based Agents","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07637","citing_title":"Learning to Communicate Locally for Large-Scale Multi-Agent Pathfinding","ref_index":98,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12491","citing_title":"Elastic Attention Cores for Scalable Vision Transformers","ref_index":124,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10537","citing_title":"Mela: Test-Time Memory Consolidation based on Transformation Hypothesis","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07637","citing_title":"Learning to Communicate Locally for Large-Scale Multi-Agent Pathfinding","ref_index":98,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21999","citing_title":"Universal Transformers Need Memory: Depth-State Trade-offs in Adaptive Recursive Reasoning","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4FWAI53MOHIDRGTQQR7VJTUV52","json":"https://pith.science/pith/4FWAI53MOHIDRGTQQR7VJTUV52.json","graph_json":"https://pith.science/api/pith-number/4FWAI53MOHIDRGTQQR7VJTUV52/graph.json","events_json":"https://pith.science/api/pith-number/4FWAI53MOHIDRGTQQR7VJTUV52/events.json","paper":"https://pith.science/paper/4FWAI53M"},"agent_actions":{"view_html":"https://pith.science/pith/4FWAI53MOHIDRGTQQR7VJTUV52","download_json":"https://pith.science/pith/4FWAI53MOHIDRGTQQR7VJTUV52.json","view_paper":"https://pith.science/paper/4FWAI53M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.11527&json=true","fetch_graph":"https://pith.science/api/pith-number/4FWAI53MOHIDRGTQQR7VJTUV52/graph.json","fetch_events":"https://pith.science/api/pith-number/4FWAI53MOHIDRGTQQR7VJTUV52/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4FWAI53MOHIDRGTQQR7VJTUV52/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4FWAI53MOHIDRGTQQR7VJTUV52/action/storage_attestation","attest_author":"https://pith.science/pith/4FWAI53MOHIDRGTQQR7VJTUV52/action/author_attestation","sign_citation":"https://pith.science/pith/4FWAI53MOHIDRGTQQR7VJTUV52/action/citation_signature","submit_replication":"https://pith.science/pith/4FWAI53MOHIDRGTQQR7VJTUV52/action/replication_record"}},"created_at":"2026-07-05T02:15:25.159354+00:00","updated_at":"2026-07-05T02:15:25.159354+00:00"}