{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ST2WLR7EIJNDAI2SG7M4AGXACY","short_pith_number":"pith:ST2WLR7E","schema_version":"1.0","canonical_sha256":"94f565c7e4425a30235237d9c01ae0160ccc30874ee649ec9b33b02285a18e2f","source":{"kind":"arxiv","id":"2411.07501","version":4},"attestation_state":"computed","paper":{"title":"LAuReL: Learned Augmented Residual Layer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Gaurav Menghani, Ravi Kumar, Sanjiv Kumar","submitted_at":"2024-11-12T02:57:15Z","abstract_excerpt":"One of the core pillars of efficient deep learning methods is architectural improvements such as the residual/skip connection, which has led to significantly better model convergence and quality. Since then the residual connection has become ubiquitous in not just convolutional neural networks but also transformer-based architectures, the backbone of LLMs.\n  In this paper we introduce Learned Augmented Residual Layer (LAuReL) -- a novel generalization of the canonical residual connection -- with the goal to be an in-situ replacement of the latter while outperforming on both model quality and f"},"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":"2411.07501","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-11-12T02:57:15Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"5e4a674482200cdd8e87e16b4abdebda28ddce2189be68148d6f4f843b455203","abstract_canon_sha256":"0df59a42f2feb5ed63d3d18922859aaf51d9250f16c061f74ee1a048d23768a7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:26:01.616157Z","signature_b64":"o6WL9Km0Pfk8VuO+d1DxtHtmAE3WobMPg6TkTuZF1DcMCo9kUzW680FoXjw0gWR9nyRuM+hC7rQ45V3RiN+LAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"94f565c7e4425a30235237d9c01ae0160ccc30874ee649ec9b33b02285a18e2f","last_reissued_at":"2026-07-05T11:26:01.615615Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:26:01.615615Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LAuReL: Learned Augmented Residual Layer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Gaurav Menghani, Ravi Kumar, Sanjiv Kumar","submitted_at":"2024-11-12T02:57:15Z","abstract_excerpt":"One of the core pillars of efficient deep learning methods is architectural improvements such as the residual/skip connection, which has led to significantly better model convergence and quality. Since then the residual connection has become ubiquitous in not just convolutional neural networks but also transformer-based architectures, the backbone of LLMs.\n  In this paper we introduce Learned Augmented Residual Layer (LAuReL) -- a novel generalization of the canonical residual connection -- with the goal to be an in-situ replacement of the latter while outperforming on both model quality and f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.07501","kind":"arxiv","version":4},"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/2411.07501/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":"2411.07501","created_at":"2026-07-05T11:26:01.615672+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.07501v4","created_at":"2026-07-05T11:26:01.615672+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.07501","created_at":"2026-07-05T11:26:01.615672+00:00"},{"alias_kind":"pith_short_12","alias_value":"ST2WLR7EIJND","created_at":"2026-07-05T11:26:01.615672+00:00"},{"alias_kind":"pith_short_16","alias_value":"ST2WLR7EIJNDAI2S","created_at":"2026-07-05T11:26:01.615672+00:00"},{"alias_kind":"pith_short_8","alias_value":"ST2WLR7E","created_at":"2026-07-05T11:26:01.615672+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05014","citing_title":"Depth-Attention: Cross-Layer Value Mixing for Language Models","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2603.15031","citing_title":"Attention Residuals","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2507.05387","citing_title":"The Generalization Ridge: Information Flow in Natural Language Generation","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03953","citing_title":"Transformers with Selective Access to Early Representations","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03953","citing_title":"Transformers with Selective Access to Early Representations","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ST2WLR7EIJNDAI2SG7M4AGXACY","json":"https://pith.science/pith/ST2WLR7EIJNDAI2SG7M4AGXACY.json","graph_json":"https://pith.science/api/pith-number/ST2WLR7EIJNDAI2SG7M4AGXACY/graph.json","events_json":"https://pith.science/api/pith-number/ST2WLR7EIJNDAI2SG7M4AGXACY/events.json","paper":"https://pith.science/paper/ST2WLR7E"},"agent_actions":{"view_html":"https://pith.science/pith/ST2WLR7EIJNDAI2SG7M4AGXACY","download_json":"https://pith.science/pith/ST2WLR7EIJNDAI2SG7M4AGXACY.json","view_paper":"https://pith.science/paper/ST2WLR7E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.07501&json=true","fetch_graph":"https://pith.science/api/pith-number/ST2WLR7EIJNDAI2SG7M4AGXACY/graph.json","fetch_events":"https://pith.science/api/pith-number/ST2WLR7EIJNDAI2SG7M4AGXACY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ST2WLR7EIJNDAI2SG7M4AGXACY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ST2WLR7EIJNDAI2SG7M4AGXACY/action/storage_attestation","attest_author":"https://pith.science/pith/ST2WLR7EIJNDAI2SG7M4AGXACY/action/author_attestation","sign_citation":"https://pith.science/pith/ST2WLR7EIJNDAI2SG7M4AGXACY/action/citation_signature","submit_replication":"https://pith.science/pith/ST2WLR7EIJNDAI2SG7M4AGXACY/action/replication_record"}},"created_at":"2026-07-05T11:26:01.615672+00:00","updated_at":"2026-07-05T11:26:01.615672+00:00"}