{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IRTW7CDMAWS4CLXUUBFEYQOMEQ","short_pith_number":"pith:IRTW7CDM","schema_version":"1.0","canonical_sha256":"44676f886c05a5c12ef4a04a4c41cc24396a99053183f5b367988718b9085055","source":{"kind":"arxiv","id":"2505.11416","version":1},"attestation_state":"computed","paper":{"title":"MID-L: Matrix-Interpolated Dropout Layer with Layer-wise Neuron Selection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.NE","authors_text":"Ariane Middel, Pouya Shaeri","submitted_at":"2025-05-16T16:29:19Z","abstract_excerpt":"Modern neural networks often activate all neurons for every input, leading to unnecessary computation and inefficiency. We introduce Matrix-Interpolated Dropout Layer (MID-L), a novel module that dynamically selects and activates only the most informative neurons by interpolating between two transformation paths via a learned, input-dependent gating vector. Unlike conventional dropout or static sparsity methods, MID-L employs a differentiable Top-k masking strategy, enabling per-input adaptive computation while maintaining end-to-end differentiability. MID-L is model-agnostic and integrates se"},"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":"2505.11416","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2025-05-16T16:29:19Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"eeefcacd4e04716a4b1212dc742e59a3f97fcf14e7c5ae2a9f55cee4a3725322","abstract_canon_sha256":"6a81a084f2d1bc83c8c348a4f1c3515e5d073ad771515ad20a642bbcd5debcd2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:04:15.715232Z","signature_b64":"rOo3EZ/TOheWF3MK1EKSBMoTAWs3A2XR9HZbBsTnq7aexm5k7+hNVJ1rlAt8IMDvh7xlH5ssvoLAHS8Xg6p/Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"44676f886c05a5c12ef4a04a4c41cc24396a99053183f5b367988718b9085055","last_reissued_at":"2026-07-05T11:04:15.714670Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:04:15.714670Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MID-L: Matrix-Interpolated Dropout Layer with Layer-wise Neuron Selection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.NE","authors_text":"Ariane Middel, Pouya Shaeri","submitted_at":"2025-05-16T16:29:19Z","abstract_excerpt":"Modern neural networks often activate all neurons for every input, leading to unnecessary computation and inefficiency. We introduce Matrix-Interpolated Dropout Layer (MID-L), a novel module that dynamically selects and activates only the most informative neurons by interpolating between two transformation paths via a learned, input-dependent gating vector. Unlike conventional dropout or static sparsity methods, MID-L employs a differentiable Top-k masking strategy, enabling per-input adaptive computation while maintaining end-to-end differentiability. MID-L is model-agnostic and integrates se"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11416","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/2505.11416/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":"2505.11416","created_at":"2026-07-05T11:04:15.714736+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.11416v1","created_at":"2026-07-05T11:04:15.714736+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11416","created_at":"2026-07-05T11:04:15.714736+00:00"},{"alias_kind":"pith_short_12","alias_value":"IRTW7CDMAWS4","created_at":"2026-07-05T11:04:15.714736+00:00"},{"alias_kind":"pith_short_16","alias_value":"IRTW7CDMAWS4CLXU","created_at":"2026-07-05T11:04:15.714736+00:00"},{"alias_kind":"pith_short_8","alias_value":"IRTW7CDM","created_at":"2026-07-05T11:04:15.714736+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.11821","citing_title":"MNIST-Gen: A Modular MNIST-Style Dataset Generation Using Hierarchical Semantics, Reinforcement Learning, and Category Theory","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IRTW7CDMAWS4CLXUUBFEYQOMEQ","json":"https://pith.science/pith/IRTW7CDMAWS4CLXUUBFEYQOMEQ.json","graph_json":"https://pith.science/api/pith-number/IRTW7CDMAWS4CLXUUBFEYQOMEQ/graph.json","events_json":"https://pith.science/api/pith-number/IRTW7CDMAWS4CLXUUBFEYQOMEQ/events.json","paper":"https://pith.science/paper/IRTW7CDM"},"agent_actions":{"view_html":"https://pith.science/pith/IRTW7CDMAWS4CLXUUBFEYQOMEQ","download_json":"https://pith.science/pith/IRTW7CDMAWS4CLXUUBFEYQOMEQ.json","view_paper":"https://pith.science/paper/IRTW7CDM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.11416&json=true","fetch_graph":"https://pith.science/api/pith-number/IRTW7CDMAWS4CLXUUBFEYQOMEQ/graph.json","fetch_events":"https://pith.science/api/pith-number/IRTW7CDMAWS4CLXUUBFEYQOMEQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IRTW7CDMAWS4CLXUUBFEYQOMEQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IRTW7CDMAWS4CLXUUBFEYQOMEQ/action/storage_attestation","attest_author":"https://pith.science/pith/IRTW7CDMAWS4CLXUUBFEYQOMEQ/action/author_attestation","sign_citation":"https://pith.science/pith/IRTW7CDMAWS4CLXUUBFEYQOMEQ/action/citation_signature","submit_replication":"https://pith.science/pith/IRTW7CDMAWS4CLXUUBFEYQOMEQ/action/replication_record"}},"created_at":"2026-07-05T11:04:15.714736+00:00","updated_at":"2026-07-05T11:04:15.714736+00:00"}