{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UNYCFHA4OLO3XL6C3ODR65DRDF","short_pith_number":"pith:UNYCFHA4","schema_version":"1.0","canonical_sha256":"a370229c1c72ddbbafc2db871f74711962f8092df741a54557788eff50570023","source":{"kind":"arxiv","id":"2510.17363","version":1},"attestation_state":"computed","paper":{"title":"M2H: Multi-Task Learning with Efficient Window-Based Cross-Task Attention for Monocular Spatial Perception","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.RO"],"primary_cat":"cs.CV","authors_text":"Francesco Nex, George Vosselman, U.V.B.L Udugama","submitted_at":"2025-10-20T10:03:31Z","abstract_excerpt":"Deploying real-time spatial perception on edge devices requires efficient multi-task models that leverage complementary task information while minimizing computational overhead. This paper introduces Multi-Mono-Hydra (M2H), a novel multi-task learning framework designed for semantic segmentation and depth, edge, and surface normal estimation from a single monocular image. Unlike conventional approaches that rely on independent single-task models or shared encoder-decoder architectures, M2H introduces a Window-Based Cross-Task Attention Module that enables structured feature exchange while pres"},"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":"2510.17363","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-10-20T10:03:31Z","cross_cats_sorted":["cs.LG","cs.RO"],"title_canon_sha256":"81e1a0d471290c1d89b494e22c6b3301e6d7443dbd632a048007da16d9c3f6db","abstract_canon_sha256":"0b57c99bbf453ef5fddb9feba403d608b7ef94fa5c84baa5685f57c28a6ace9e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:15.128720Z","signature_b64":"R5ud+FysjOaZDKdmdk7KlBbZSK2j7QyNBU81l/niNT6Skuo1mD8xWR4mMZ0YrWtGMJUsfvZAFzz0ZjIWiN28Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a370229c1c72ddbbafc2db871f74711962f8092df741a54557788eff50570023","last_reissued_at":"2026-05-20T00:04:15.127937Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:15.127937Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"M2H: Multi-Task Learning with Efficient Window-Based Cross-Task Attention for Monocular Spatial Perception","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.RO"],"primary_cat":"cs.CV","authors_text":"Francesco Nex, George Vosselman, U.V.B.L Udugama","submitted_at":"2025-10-20T10:03:31Z","abstract_excerpt":"Deploying real-time spatial perception on edge devices requires efficient multi-task models that leverage complementary task information while minimizing computational overhead. This paper introduces Multi-Mono-Hydra (M2H), a novel multi-task learning framework designed for semantic segmentation and depth, edge, and surface normal estimation from a single monocular image. Unlike conventional approaches that rely on independent single-task models or shared encoder-decoder architectures, M2H introduces a Window-Based Cross-Task Attention Module that enables structured feature exchange while pres"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.17363","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/2510.17363/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":"2510.17363","created_at":"2026-05-20T00:04:15.128041+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.17363v1","created_at":"2026-05-20T00:04:15.128041+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.17363","created_at":"2026-05-20T00:04:15.128041+00:00"},{"alias_kind":"pith_short_12","alias_value":"UNYCFHA4OLO3","created_at":"2026-05-20T00:04:15.128041+00:00"},{"alias_kind":"pith_short_16","alias_value":"UNYCFHA4OLO3XL6C","created_at":"2026-05-20T00:04:15.128041+00:00"},{"alias_kind":"pith_short_8","alias_value":"UNYCFHA4","created_at":"2026-05-20T00:04:15.128041+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.17661","citing_title":"Mono-Hydra++: Real-Time Monocular Scene Graph Construction with Multi-Task Learning for 3D Indoor Mapping","ref_index":13,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UNYCFHA4OLO3XL6C3ODR65DRDF","json":"https://pith.science/pith/UNYCFHA4OLO3XL6C3ODR65DRDF.json","graph_json":"https://pith.science/api/pith-number/UNYCFHA4OLO3XL6C3ODR65DRDF/graph.json","events_json":"https://pith.science/api/pith-number/UNYCFHA4OLO3XL6C3ODR65DRDF/events.json","paper":"https://pith.science/paper/UNYCFHA4"},"agent_actions":{"view_html":"https://pith.science/pith/UNYCFHA4OLO3XL6C3ODR65DRDF","download_json":"https://pith.science/pith/UNYCFHA4OLO3XL6C3ODR65DRDF.json","view_paper":"https://pith.science/paper/UNYCFHA4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.17363&json=true","fetch_graph":"https://pith.science/api/pith-number/UNYCFHA4OLO3XL6C3ODR65DRDF/graph.json","fetch_events":"https://pith.science/api/pith-number/UNYCFHA4OLO3XL6C3ODR65DRDF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UNYCFHA4OLO3XL6C3ODR65DRDF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UNYCFHA4OLO3XL6C3ODR65DRDF/action/storage_attestation","attest_author":"https://pith.science/pith/UNYCFHA4OLO3XL6C3ODR65DRDF/action/author_attestation","sign_citation":"https://pith.science/pith/UNYCFHA4OLO3XL6C3ODR65DRDF/action/citation_signature","submit_replication":"https://pith.science/pith/UNYCFHA4OLO3XL6C3ODR65DRDF/action/replication_record"}},"created_at":"2026-05-20T00:04:15.128041+00:00","updated_at":"2026-05-20T00:04:15.128041+00:00"}