{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7FDFUTEC2KPPIHLTYOU6SCXFWJ","short_pith_number":"pith:7FDFUTEC","schema_version":"1.0","canonical_sha256":"f9465a4c82d29ef41d73c3a9e90ae5b2618263e816a11f03c7a8f07ce9259405","source":{"kind":"arxiv","id":"2504.13173","version":1},"attestation_state":"computed","paper":{"title":"It's All Connected: A Journey Through Test-Time Memorization, Attentional Bias, Retention, and Online Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ali Behrouz, Meisam Razaviyayn, Peilin Zhong, Vahab Mirrokni","submitted_at":"2025-04-17T17:59:33Z","abstract_excerpt":"Designing efficient and effective architectural backbones has been in the core of research efforts to enhance the capability of foundation models. Inspired by the human cognitive phenomenon of attentional bias-the natural tendency to prioritize certain events or stimuli-we reconceptualize neural architectures, including Transformers, Titans, and modern linear recurrent neural networks as associative memory modules that learn a mapping of keys and values using an internal objective, referred to as attentional bias. Surprisingly, we observed that most existing sequence models leverage either (1)"},"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":"2504.13173","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-17T17:59:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"97cd9d8c4c2373fdea9203b2033a72175b460822940d87658117a536a7ee6d6d","abstract_canon_sha256":"7ea0f229443a7d156141261e8117f3cc4e9c8433399fefcaf16972b0062919f3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:50:38.943832Z","signature_b64":"t6ZoiaeEMz3yaL3ZX/KvIqfbFk/JLsCY+2E1qpzDbrv+/w0BxZ9hk3V8558U1HsX/bugTr18KSdILkf4akS2DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f9465a4c82d29ef41d73c3a9e90ae5b2618263e816a11f03c7a8f07ce9259405","last_reissued_at":"2026-07-05T10:50:38.943342Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:50:38.943342Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"It's All Connected: A Journey Through Test-Time Memorization, Attentional Bias, Retention, and Online Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ali Behrouz, Meisam Razaviyayn, Peilin Zhong, Vahab Mirrokni","submitted_at":"2025-04-17T17:59:33Z","abstract_excerpt":"Designing efficient and effective architectural backbones has been in the core of research efforts to enhance the capability of foundation models. Inspired by the human cognitive phenomenon of attentional bias-the natural tendency to prioritize certain events or stimuli-we reconceptualize neural architectures, including Transformers, Titans, and modern linear recurrent neural networks as associative memory modules that learn a mapping of keys and values using an internal objective, referred to as attentional bias. Surprisingly, we observed that most existing sequence models leverage either (1)"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.13173","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/2504.13173/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":"2504.13173","created_at":"2026-07-05T10:50:38.943414+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.13173v1","created_at":"2026-07-05T10:50:38.943414+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.13173","created_at":"2026-07-05T10:50:38.943414+00:00"},{"alias_kind":"pith_short_12","alias_value":"7FDFUTEC2KPP","created_at":"2026-07-05T10:50:38.943414+00:00"},{"alias_kind":"pith_short_16","alias_value":"7FDFUTEC2KPPIHLT","created_at":"2026-07-05T10:50:38.943414+00:00"},{"alias_kind":"pith_short_8","alias_value":"7FDFUTEC","created_at":"2026-07-05T10:50:38.943414+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25342","citing_title":"Lifelong In-Context Learning with Transformers Requires Parametric Forms of Attention","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02775","citing_title":"AURA: Action-Gated Memory for Robot Policies at Constant VRAM","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25949","citing_title":"Small Models, Strong Priors: Architectural Inductive Bias for Parameter-Efficient Neural PDE Solvers","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28773","citing_title":"Rethinking Memory as Continuously Evolving Connectivity","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19403","citing_title":"TIDE: Asymmetric Neural Circuits for Stabilized Temporal Inhibitory-Excitatory Dynamics","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2509.22321","citing_title":"Distributed Associative Memory via Online Convex Optimization","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2509.26645","citing_title":"TTT3R: 3D Reconstruction as Test-Time Training","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2505.23884","citing_title":"Test-Time Training Done Right","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2603.23516","citing_title":"MSA: Memory Sparse Attention for Efficient End-to-End Memory Model Scaling to 100M Tokens","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00292","citing_title":"Caracal: Causal Architecture via Spectral Mixing","ref_index":108,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08542","citing_title":"Scal3R: Scalable Test-Time Training for Large-Scale 3D Reconstruction","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07350","citing_title":"Fast Spatial Memory with Elastic Test-Time Training","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06169","citing_title":"In-Place Test-Time Training","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7FDFUTEC2KPPIHLTYOU6SCXFWJ","json":"https://pith.science/pith/7FDFUTEC2KPPIHLTYOU6SCXFWJ.json","graph_json":"https://pith.science/api/pith-number/7FDFUTEC2KPPIHLTYOU6SCXFWJ/graph.json","events_json":"https://pith.science/api/pith-number/7FDFUTEC2KPPIHLTYOU6SCXFWJ/events.json","paper":"https://pith.science/paper/7FDFUTEC"},"agent_actions":{"view_html":"https://pith.science/pith/7FDFUTEC2KPPIHLTYOU6SCXFWJ","download_json":"https://pith.science/pith/7FDFUTEC2KPPIHLTYOU6SCXFWJ.json","view_paper":"https://pith.science/paper/7FDFUTEC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.13173&json=true","fetch_graph":"https://pith.science/api/pith-number/7FDFUTEC2KPPIHLTYOU6SCXFWJ/graph.json","fetch_events":"https://pith.science/api/pith-number/7FDFUTEC2KPPIHLTYOU6SCXFWJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7FDFUTEC2KPPIHLTYOU6SCXFWJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7FDFUTEC2KPPIHLTYOU6SCXFWJ/action/storage_attestation","attest_author":"https://pith.science/pith/7FDFUTEC2KPPIHLTYOU6SCXFWJ/action/author_attestation","sign_citation":"https://pith.science/pith/7FDFUTEC2KPPIHLTYOU6SCXFWJ/action/citation_signature","submit_replication":"https://pith.science/pith/7FDFUTEC2KPPIHLTYOU6SCXFWJ/action/replication_record"}},"created_at":"2026-07-05T10:50:38.943414+00:00","updated_at":"2026-07-05T10:50:38.943414+00:00"}