{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6MTMXO4YWNDK5NBWHP3ADWOZMM","short_pith_number":"pith:6MTMXO4Y","schema_version":"1.0","canonical_sha256":"f326cbbb98b346aeb4363bf601d9d9631e61a013590b9a1be1db1d8f5d1beb62","source":{"kind":"arxiv","id":"2505.17909","version":1},"attestation_state":"computed","paper":{"title":"NeuroTrails: Training with Dynamic Sparse Heads as the Key to Effective Ensembling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Boqian Wu, Bram Grooten, Chenxiang Zhang, Decebal Constantin Mocanu, Elena Mocanu, Farid Hasanov, Ghada Sokar, Lu Yin, Mykola Pechenizkiy, Qiao Xiao, Shiwei Liu, Zahra Atashgahi","submitted_at":"2025-05-23T13:53:21Z","abstract_excerpt":"Model ensembles have long been a cornerstone for improving generalization and robustness in deep learning. However, their effectiveness often comes at the cost of substantial computational overhead. To address this issue, state-of-the-art methods aim to replicate ensemble-class performance without requiring multiple independently trained networks. Unfortunately, these algorithms often still demand considerable compute at inference. In response to these limitations, we introduce $\\textbf{NeuroTrails}$, a sparse multi-head architecture with dynamically evolving topology. This unexplored model-ag"},"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":"2505.17909","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T13:53:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f42a4726ccd35d5a41236fab1f7e364bfd92e6a6482b785cacdf2a92133627fc","abstract_canon_sha256":"76d076eb490d4521857cf72385dbded1e622d1ac258959b3f427fc0b0165fe4b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:33.667533Z","signature_b64":"dIQ39ji94nPa4fttJtbkdd0YNwg8T7NpTom19twQsWQcvUq2Pq7JK1rs2EnJ8Y8f04odPDRcW9XLxegBZQtgDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f326cbbb98b346aeb4363bf601d9d9631e61a013590b9a1be1db1d8f5d1beb62","last_reissued_at":"2026-07-05T11:08:33.666962Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:33.666962Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"NeuroTrails: Training with Dynamic Sparse Heads as the Key to Effective Ensembling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Boqian Wu, Bram Grooten, Chenxiang Zhang, Decebal Constantin Mocanu, Elena Mocanu, Farid Hasanov, Ghada Sokar, Lu Yin, Mykola Pechenizkiy, Qiao Xiao, Shiwei Liu, Zahra Atashgahi","submitted_at":"2025-05-23T13:53:21Z","abstract_excerpt":"Model ensembles have long been a cornerstone for improving generalization and robustness in deep learning. However, their effectiveness often comes at the cost of substantial computational overhead. To address this issue, state-of-the-art methods aim to replicate ensemble-class performance without requiring multiple independently trained networks. Unfortunately, these algorithms often still demand considerable compute at inference. In response to these limitations, we introduce $\\textbf{NeuroTrails}$, a sparse multi-head architecture with dynamically evolving topology. This unexplored model-ag"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.17909","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/2505.17909/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":"2505.17909","created_at":"2026-07-05T11:08:33.667032+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.17909v1","created_at":"2026-07-05T11:08:33.667032+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.17909","created_at":"2026-07-05T11:08:33.667032+00:00"},{"alias_kind":"pith_short_12","alias_value":"6MTMXO4YWNDK","created_at":"2026-07-05T11:08:33.667032+00:00"},{"alias_kind":"pith_short_16","alias_value":"6MTMXO4YWNDK5NBW","created_at":"2026-07-05T11:08:33.667032+00:00"},{"alias_kind":"pith_short_8","alias_value":"6MTMXO4Y","created_at":"2026-07-05T11:08:33.667032+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6MTMXO4YWNDK5NBWHP3ADWOZMM","json":"https://pith.science/pith/6MTMXO4YWNDK5NBWHP3ADWOZMM.json","graph_json":"https://pith.science/api/pith-number/6MTMXO4YWNDK5NBWHP3ADWOZMM/graph.json","events_json":"https://pith.science/api/pith-number/6MTMXO4YWNDK5NBWHP3ADWOZMM/events.json","paper":"https://pith.science/paper/6MTMXO4Y"},"agent_actions":{"view_html":"https://pith.science/pith/6MTMXO4YWNDK5NBWHP3ADWOZMM","download_json":"https://pith.science/pith/6MTMXO4YWNDK5NBWHP3ADWOZMM.json","view_paper":"https://pith.science/paper/6MTMXO4Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.17909&json=true","fetch_graph":"https://pith.science/api/pith-number/6MTMXO4YWNDK5NBWHP3ADWOZMM/graph.json","fetch_events":"https://pith.science/api/pith-number/6MTMXO4YWNDK5NBWHP3ADWOZMM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6MTMXO4YWNDK5NBWHP3ADWOZMM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6MTMXO4YWNDK5NBWHP3ADWOZMM/action/storage_attestation","attest_author":"https://pith.science/pith/6MTMXO4YWNDK5NBWHP3ADWOZMM/action/author_attestation","sign_citation":"https://pith.science/pith/6MTMXO4YWNDK5NBWHP3ADWOZMM/action/citation_signature","submit_replication":"https://pith.science/pith/6MTMXO4YWNDK5NBWHP3ADWOZMM/action/replication_record"}},"created_at":"2026-07-05T11:08:33.667032+00:00","updated_at":"2026-07-05T11:08:33.667032+00:00"}