{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7NE5AXAGLVSKYM3GBRSRRU5SXH","short_pith_number":"pith:7NE5AXAG","schema_version":"1.0","canonical_sha256":"fb49d05c065d64ac33660c6518d3b2b9db4325421db3d99f0b0354742c968b79","source":{"kind":"arxiv","id":"2508.07315","version":2},"attestation_state":"computed","paper":{"title":"FlexCTC: GPU-powered CTC Beam Decoding With Advanced Contextual Abilities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Andrei Andrusenko, Boris Ginsburg, Lilit Grigoryan, Nikolay Karpov, Vitaly Lavrukhin, Vladimir Bataev","submitted_at":"2025-08-10T12:15:57Z","abstract_excerpt":"While beam search improves speech recognition quality over greedy decoding, standard implementations are slow, often sequential, and CPU-bound. To fully leverage modern hardware capabilities, we present a novel open-source FlexCTC toolkit for fully GPU-based beam decoding, designed for Connectionist Temporal Classification (CTC) models. Developed entirely in Python and PyTorch, it offers a fast, user-friendly, and extensible alternative to traditional C++, CUDA, or WFST-based decoders. The toolkit features a high-performance, fully batched GPU implementation with eliminated CPU-GPU synchroniza"},"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":"2508.07315","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2025-08-10T12:15:57Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG","cs.SD"],"title_canon_sha256":"003779d9950fe78e398564a206d3d647847bbf707e5ace14353f3c433e243d79","abstract_canon_sha256":"da8e2d00644f6709842b0b0c5671d1dd8c69dcf0462722d9256d63c5606aa9f8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:07.435850Z","signature_b64":"Ph/YPnl9CSX3/FMI0ZJRzKec/yGSeyMxGpz8ls1+DZhCps14bUwC6tLKC6i0gZN+FEYSI7nPMEnu3eSDdQ7lCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fb49d05c065d64ac33660c6518d3b2b9db4325421db3d99f0b0354742c968b79","last_reissued_at":"2026-07-05T11:53:07.435382Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:07.435382Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FlexCTC: GPU-powered CTC Beam Decoding With Advanced Contextual Abilities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Andrei Andrusenko, Boris Ginsburg, Lilit Grigoryan, Nikolay Karpov, Vitaly Lavrukhin, Vladimir Bataev","submitted_at":"2025-08-10T12:15:57Z","abstract_excerpt":"While beam search improves speech recognition quality over greedy decoding, standard implementations are slow, often sequential, and CPU-bound. To fully leverage modern hardware capabilities, we present a novel open-source FlexCTC toolkit for fully GPU-based beam decoding, designed for Connectionist Temporal Classification (CTC) models. Developed entirely in Python and PyTorch, it offers a fast, user-friendly, and extensible alternative to traditional C++, CUDA, or WFST-based decoders. The toolkit features a high-performance, fully batched GPU implementation with eliminated CPU-GPU synchroniza"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.07315","kind":"arxiv","version":2},"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/2508.07315/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":"2508.07315","created_at":"2026-07-05T11:53:07.435437+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.07315v2","created_at":"2026-07-05T11:53:07.435437+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07315","created_at":"2026-07-05T11:53:07.435437+00:00"},{"alias_kind":"pith_short_12","alias_value":"7NE5AXAGLVSK","created_at":"2026-07-05T11:53:07.435437+00:00"},{"alias_kind":"pith_short_16","alias_value":"7NE5AXAGLVSKYM3G","created_at":"2026-07-05T11:53:07.435437+00:00"},{"alias_kind":"pith_short_8","alias_value":"7NE5AXAG","created_at":"2026-07-05T11:53:07.435437+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.07354","citing_title":"Contextual Earnings-22: A Speech Recognition Benchmark with Custom Vocabulary in the Wild","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7NE5AXAGLVSKYM3GBRSRRU5SXH","json":"https://pith.science/pith/7NE5AXAGLVSKYM3GBRSRRU5SXH.json","graph_json":"https://pith.science/api/pith-number/7NE5AXAGLVSKYM3GBRSRRU5SXH/graph.json","events_json":"https://pith.science/api/pith-number/7NE5AXAGLVSKYM3GBRSRRU5SXH/events.json","paper":"https://pith.science/paper/7NE5AXAG"},"agent_actions":{"view_html":"https://pith.science/pith/7NE5AXAGLVSKYM3GBRSRRU5SXH","download_json":"https://pith.science/pith/7NE5AXAGLVSKYM3GBRSRRU5SXH.json","view_paper":"https://pith.science/paper/7NE5AXAG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.07315&json=true","fetch_graph":"https://pith.science/api/pith-number/7NE5AXAGLVSKYM3GBRSRRU5SXH/graph.json","fetch_events":"https://pith.science/api/pith-number/7NE5AXAGLVSKYM3GBRSRRU5SXH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7NE5AXAGLVSKYM3GBRSRRU5SXH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7NE5AXAGLVSKYM3GBRSRRU5SXH/action/storage_attestation","attest_author":"https://pith.science/pith/7NE5AXAGLVSKYM3GBRSRRU5SXH/action/author_attestation","sign_citation":"https://pith.science/pith/7NE5AXAGLVSKYM3GBRSRRU5SXH/action/citation_signature","submit_replication":"https://pith.science/pith/7NE5AXAGLVSKYM3GBRSRRU5SXH/action/replication_record"}},"created_at":"2026-07-05T11:53:07.435437+00:00","updated_at":"2026-07-05T11:53:07.435437+00:00"}