{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:JV4XQSVUIJIKGP7FBHTCFFMXTS","short_pith_number":"pith:JV4XQSVU","schema_version":"1.0","canonical_sha256":"4d79784ab44250a33fe509e62295979cbacac1299fc189bad60603c0ffc60ceb","source":{"kind":"arxiv","id":"2307.04552","version":1},"attestation_state":"computed","paper":{"title":"SparseVSR: Lightweight and Noise Robust Visual Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Adriana Fernandez-Lopez, Alexandros Haliassos, Honglie Chen, Maja Pantic, Pingchuan Ma, Stavros Petridis","submitted_at":"2023-07-10T13:34:13Z","abstract_excerpt":"Recent advances in deep neural networks have achieved unprecedented success in visual speech recognition. However, there remains substantial disparity between current methods and their deployment in resource-constrained devices. In this work, we explore different magnitude-based pruning techniques to generate a lightweight model that achieves higher performance than its dense model equivalent, especially under the presence of visual noise. Our sparse models achieve state-of-the-art results at 10% sparsity on the LRS3 dataset and outperform the dense equivalent up to 70% sparsity. We evaluate o"},"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":"2307.04552","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-07-10T13:34:13Z","cross_cats_sorted":[],"title_canon_sha256":"43a665c9eef4708b424a35fb9cac04b556c2f1edd6f37aa97c3fd9d6f03d82e4","abstract_canon_sha256":"d9c1951f036f56aa6a4471a4b617550ef1d91f4e49f7792dd0712ea7e76319d0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:29:14.928498Z","signature_b64":"2daOJ6Ku+PEgN8dIfpLYvZutulq6NZTfdaTRtKXC+20gvqqZND5DAM3ha3l6OINXlPaIi6MJ3ZtxF7tNpowAAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4d79784ab44250a33fe509e62295979cbacac1299fc189bad60603c0ffc60ceb","last_reissued_at":"2026-07-05T06:29:14.928071Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:29:14.928071Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SparseVSR: Lightweight and Noise Robust Visual Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Adriana Fernandez-Lopez, Alexandros Haliassos, Honglie Chen, Maja Pantic, Pingchuan Ma, Stavros Petridis","submitted_at":"2023-07-10T13:34:13Z","abstract_excerpt":"Recent advances in deep neural networks have achieved unprecedented success in visual speech recognition. However, there remains substantial disparity between current methods and their deployment in resource-constrained devices. In this work, we explore different magnitude-based pruning techniques to generate a lightweight model that achieves higher performance than its dense model equivalent, especially under the presence of visual noise. Our sparse models achieve state-of-the-art results at 10% sparsity on the LRS3 dataset and outperform the dense equivalent up to 70% sparsity. We evaluate o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.04552","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/2307.04552/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":"2307.04552","created_at":"2026-07-05T06:29:14.928126+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.04552v1","created_at":"2026-07-05T06:29:14.928126+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.04552","created_at":"2026-07-05T06:29:14.928126+00:00"},{"alias_kind":"pith_short_12","alias_value":"JV4XQSVUIJIK","created_at":"2026-07-05T06:29:14.928126+00:00"},{"alias_kind":"pith_short_16","alias_value":"JV4XQSVUIJIKGP7F","created_at":"2026-07-05T06:29:14.928126+00:00"},{"alias_kind":"pith_short_8","alias_value":"JV4XQSVU","created_at":"2026-07-05T06:29:14.928126+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00751","citing_title":"Head-Pose-Aware Visual Speech Recognition with FiLM Modulation","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2310.02277","citing_title":"Junk DNA Hypothesis: Pruning Small Pre-Trained Weights Irreversibly and Monotonically Impairs \"Difficult\" Downstream Tasks in LLMs","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JV4XQSVUIJIKGP7FBHTCFFMXTS","json":"https://pith.science/pith/JV4XQSVUIJIKGP7FBHTCFFMXTS.json","graph_json":"https://pith.science/api/pith-number/JV4XQSVUIJIKGP7FBHTCFFMXTS/graph.json","events_json":"https://pith.science/api/pith-number/JV4XQSVUIJIKGP7FBHTCFFMXTS/events.json","paper":"https://pith.science/paper/JV4XQSVU"},"agent_actions":{"view_html":"https://pith.science/pith/JV4XQSVUIJIKGP7FBHTCFFMXTS","download_json":"https://pith.science/pith/JV4XQSVUIJIKGP7FBHTCFFMXTS.json","view_paper":"https://pith.science/paper/JV4XQSVU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.04552&json=true","fetch_graph":"https://pith.science/api/pith-number/JV4XQSVUIJIKGP7FBHTCFFMXTS/graph.json","fetch_events":"https://pith.science/api/pith-number/JV4XQSVUIJIKGP7FBHTCFFMXTS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JV4XQSVUIJIKGP7FBHTCFFMXTS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JV4XQSVUIJIKGP7FBHTCFFMXTS/action/storage_attestation","attest_author":"https://pith.science/pith/JV4XQSVUIJIKGP7FBHTCFFMXTS/action/author_attestation","sign_citation":"https://pith.science/pith/JV4XQSVUIJIKGP7FBHTCFFMXTS/action/citation_signature","submit_replication":"https://pith.science/pith/JV4XQSVUIJIKGP7FBHTCFFMXTS/action/replication_record"}},"created_at":"2026-07-05T06:29:14.928126+00:00","updated_at":"2026-07-05T06:29:14.928126+00:00"}