{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:RWPCTPJLWPFBOS2ZUN22NCRY65","short_pith_number":"pith:RWPCTPJL","schema_version":"1.0","canonical_sha256":"8d9e29bd2bb3ca174b59a375a68a38f75a8fe70f86fcdbff0c7e6533c7559edb","source":{"kind":"arxiv","id":"2605.14705","version":1},"attestation_state":"computed","paper":{"title":"Towards Continuous Sign Language Conversation from Isolated Signs","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"SignaVox generates 3D sign language responses directly from prior signing context without text or glosses.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chanyoung Kim, Jiwoo Park, Junhyeok Kim, Kyobin Choo, Minseo Kim, Seong Jae Hwang, Youngmin Kim","submitted_at":"2026-05-14T11:22:27Z","abstract_excerpt":"Sign language is the primary language for many Deaf and Hard-of-Hearing (DHH) signers, yet most conversational AI systems still mediate interaction through spoken or written language. This spoken-language-centered interface can limit access for signers for whom spoken or written language is not the most accessible medium, motivating direct sign-to-sign conversational modeling. However, sentence-level sign video data are expensive to collect and annotate, leaving existing sign translation and production models with limited vocabulary coverage and weak open-domain generalization. We address this"},"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":true,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.14705","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T11:22:27Z","cross_cats_sorted":[],"title_canon_sha256":"3787827a34fe7a7499ea1aab7ff42a63fe1a511ff566e18dac4dbc97214d9b6d","abstract_canon_sha256":"e0b60c4cf692b6cc64e2e598144604ef4cb967219f8df0a8eac84ec4dae19222"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:59.278832Z","signature_b64":"IC84XBD8tvW1FhMkciCfd/kAh7oPRWNZ10ht0WDKZidz/SW+Jgmk6YTivFWCxFlrJrf7RSm4Cco8d08gmfr2AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8d9e29bd2bb3ca174b59a375a68a38f75a8fe70f86fcdbff0c7e6533c7559edb","last_reissued_at":"2026-05-17T23:38:59.278087Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:59.278087Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Continuous Sign Language Conversation from Isolated Signs","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"SignaVox generates 3D sign language responses directly from prior signing context without text or glosses.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chanyoung Kim, Jiwoo Park, Junhyeok Kim, Kyobin Choo, Minseo Kim, Seong Jae Hwang, Youngmin Kim","submitted_at":"2026-05-14T11:22:27Z","abstract_excerpt":"Sign language is the primary language for many Deaf and Hard-of-Hearing (DHH) signers, yet most conversational AI systems still mediate interaction through spoken or written language. This spoken-language-centered interface can limit access for signers for whom spoken or written language is not the most accessible medium, motivating direct sign-to-sign conversational modeling. However, sentence-level sign video data are expensive to collect and annotate, leaving existing sign translation and production models with limited vocabulary coverage and weak open-domain generalization. We address this"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We train SignaVox, a direct sign-to-sign conversational model that generates 3D body, hand, and facial motion responses from prior signing context without spoken-language text or externally provided glosses at inference time.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the recomposed continuous sign videos from isolated clips using BRAID accurately capture natural co-articulation and semantics, and that the retrieval-guided translator provides high-quality gloss sequences without significant errors.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Constructs continuous sign conversation data from isolated signs using retrieval and diffusion models to train a direct sign-to-sign conversational AI.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"SignaVox generates 3D sign language responses directly from prior signing context without text or glosses.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"cecea1a96548a73ef2a049c343bc3abbc83c2df3f9a2a88eb7b7b0fc8ed94940"},"source":{"id":"2605.14705","kind":"arxiv","version":1},"verdict":{"id":"cb0e69a4-6ad0-4f7d-8470-39b0db41abe1","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T04:55:46.619624Z","strongest_claim":"We train SignaVox, a direct sign-to-sign conversational model that generates 3D body, hand, and facial motion responses from prior signing context without spoken-language text or externally provided glosses at inference time.","one_line_summary":"Constructs continuous sign conversation data from isolated signs using retrieval and diffusion models to train a direct sign-to-sign conversational AI.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the recomposed continuous sign videos from isolated clips using BRAID accurately capture natural co-articulation and semantics, and that the retrieval-guided translator provides high-quality gloss sequences without significant errors.","pith_extraction_headline":"SignaVox generates 3D sign language responses directly from prior signing context without text or glosses."},"references":{"count":112,"sample":[{"doi":"","year":2020,"title":"Bsl-1k: Scaling up co-articulated sign language recognition using mouthing cues","work_id":"60383154-e5af-4344-9ede-5a051760ae08","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2008,"title":"The american sign language lexicon video dataset","work_id":"1394910f-f5b9-454d-b11f-32ee3c232fec","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Qwen3-VL Technical Report","work_id":"1fe243aa-e3c0-4da6-b391-4cbcfc88d5c0","ref_index":3,"cited_arxiv_id":"2511.21631","is_internal_anchor":true},{"doi":"","year":2024,"title":"Parent american sign language skills correlate with child–but not toddler–asl vocabulary size.Language Acquisition, 31(2):85–99, 2024","work_id":"ebf9ce86-ff50-473a-b0b8-99ea220bec7e","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"Sign language recognition, generation, and translation: An interdisciplinary perspective","work_id":"5bc0e8be-e9d6-4dfd-834c-a2f6ea3d273d","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":112,"snapshot_sha256":"f2689de3017bfd0466407270cb21b94e607b9db57d473f0d3de5a652e8d9186b","internal_anchors":5},"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":"2605.14705","created_at":"2026-05-17T23:38:59.278195+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.14705v1","created_at":"2026-05-17T23:38:59.278195+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14705","created_at":"2026-05-17T23:38:59.278195+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/RWPCTPJLWPFBOS2ZUN22NCRY65","json":"https://pith.science/pith/RWPCTPJLWPFBOS2ZUN22NCRY65.json","graph_json":"https://pith.science/api/pith-number/RWPCTPJLWPFBOS2ZUN22NCRY65/graph.json","events_json":"https://pith.science/api/pith-number/RWPCTPJLWPFBOS2ZUN22NCRY65/events.json","paper":"https://pith.science/paper/RWPCTPJL"},"agent_actions":{"view_html":"https://pith.science/pith/RWPCTPJLWPFBOS2ZUN22NCRY65","download_json":"https://pith.science/pith/RWPCTPJLWPFBOS2ZUN22NCRY65.json","view_paper":"https://pith.science/paper/RWPCTPJL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.14705&json=true","fetch_graph":"https://pith.science/api/pith-number/RWPCTPJLWPFBOS2ZUN22NCRY65/graph.json","fetch_events":"https://pith.science/api/pith-number/RWPCTPJLWPFBOS2ZUN22NCRY65/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RWPCTPJLWPFBOS2ZUN22NCRY65/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RWPCTPJLWPFBOS2ZUN22NCRY65/action/storage_attestation","attest_author":"https://pith.science/pith/RWPCTPJLWPFBOS2ZUN22NCRY65/action/author_attestation","sign_citation":"https://pith.science/pith/RWPCTPJLWPFBOS2ZUN22NCRY65/action/citation_signature","submit_replication":"https://pith.science/pith/RWPCTPJLWPFBOS2ZUN22NCRY65/action/replication_record"}},"created_at":"2026-05-17T23:38:59.278195+00:00","updated_at":"2026-05-17T23:38:59.278195+00:00"}