{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SLFF4WIDEHVUERWAEBSP2QPPK6","short_pith_number":"pith:SLFF4WID","schema_version":"1.0","canonical_sha256":"92ca5e590321eb4246c02064fd41ef579cf93d8cb0d003616717c3513188b73e","source":{"kind":"arxiv","id":"2608.00235","version":1},"attestation_state":"computed","paper":{"title":"Attention-Steered Vision-Language Models for Sign Language Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Annemarie D. Ross, Guohao Sun, Meibo Hu, Sheng Li, Zhiqiang Tao","submitted_at":"2026-07-31T19:30:03Z","abstract_excerpt":"Vision-language models (VLMs) have emerged as a powerful framework for multimodal video understanding. However, they remain limited in the sign language translation task, where we identify a key failure mode of existing VLMbased translators: poor spatial-temporal visual grounding. In particular, we find that standard next-token cross-entropy does not directly provide signal for where and when the model should attend, causing models to overlook sign-relevant regions and frames. To address this challenge, we propose AttnSign, a VLM-based spatial-temporal attention steering framework for sign lan"},"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":"2608.00235","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-07-31T19:30:03Z","cross_cats_sorted":[],"title_canon_sha256":"c415eac432d9c30c382a91bae877910a08633a59ea50a4d3fbea16644da6b941","abstract_canon_sha256":"56f0aca3def6f8797af77a57ed78010655f2767596b3f61a7046c778741c3c66"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T00:34:27.217262Z","signature_b64":"596PWdGdcQ/g7QFoHKdzlVcNgFU5RZ38N68fmDQQAg1deY1GIEaKycXqQ/CV/GFwCfAprc6eCGvemKe7VnuOAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"92ca5e590321eb4246c02064fd41ef579cf93d8cb0d003616717c3513188b73e","last_reissued_at":"2026-08-04T00:34:27.215806Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T00:34:27.215806Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Attention-Steered Vision-Language Models for Sign Language Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Annemarie D. Ross, Guohao Sun, Meibo Hu, Sheng Li, Zhiqiang Tao","submitted_at":"2026-07-31T19:30:03Z","abstract_excerpt":"Vision-language models (VLMs) have emerged as a powerful framework for multimodal video understanding. However, they remain limited in the sign language translation task, where we identify a key failure mode of existing VLMbased translators: poor spatial-temporal visual grounding. In particular, we find that standard next-token cross-entropy does not directly provide signal for where and when the model should attend, causing models to overlook sign-relevant regions and frames. To address this challenge, we propose AttnSign, a VLM-based spatial-temporal attention steering framework for sign lan"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.00235","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/2608.00235/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":"2608.00235","created_at":"2026-08-04T00:34:27.216882+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.00235v1","created_at":"2026-08-04T00:34:27.216882+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.00235","created_at":"2026-08-04T00:34:27.216882+00:00"},{"alias_kind":"pith_short_12","alias_value":"SLFF4WIDEHVU","created_at":"2026-08-04T00:34:27.216882+00:00"},{"alias_kind":"pith_short_16","alias_value":"SLFF4WIDEHVUERWA","created_at":"2026-08-04T00:34:27.216882+00:00"},{"alias_kind":"pith_short_8","alias_value":"SLFF4WID","created_at":"2026-08-04T00:34:27.216882+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/SLFF4WIDEHVUERWAEBSP2QPPK6","json":"https://pith.science/pith/SLFF4WIDEHVUERWAEBSP2QPPK6.json","graph_json":"https://pith.science/api/pith-number/SLFF4WIDEHVUERWAEBSP2QPPK6/graph.json","events_json":"https://pith.science/api/pith-number/SLFF4WIDEHVUERWAEBSP2QPPK6/events.json","paper":"https://pith.science/paper/SLFF4WID"},"agent_actions":{"view_html":"https://pith.science/pith/SLFF4WIDEHVUERWAEBSP2QPPK6","download_json":"https://pith.science/pith/SLFF4WIDEHVUERWAEBSP2QPPK6.json","view_paper":"https://pith.science/paper/SLFF4WID","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.00235&json=true","fetch_graph":"https://pith.science/api/pith-number/SLFF4WIDEHVUERWAEBSP2QPPK6/graph.json","fetch_events":"https://pith.science/api/pith-number/SLFF4WIDEHVUERWAEBSP2QPPK6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SLFF4WIDEHVUERWAEBSP2QPPK6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SLFF4WIDEHVUERWAEBSP2QPPK6/action/storage_attestation","attest_author":"https://pith.science/pith/SLFF4WIDEHVUERWAEBSP2QPPK6/action/author_attestation","sign_citation":"https://pith.science/pith/SLFF4WIDEHVUERWAEBSP2QPPK6/action/citation_signature","submit_replication":"https://pith.science/pith/SLFF4WIDEHVUERWAEBSP2QPPK6/action/replication_record"}},"created_at":"2026-08-04T00:34:27.216882+00:00","updated_at":"2026-08-04T00:34:27.216882+00:00"}