{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZEV5662LEFUZP6CJLI555G2PLD","short_pith_number":"pith:ZEV5662L","schema_version":"1.0","canonical_sha256":"c92bdf7b4b216997f8495a3bde9b4f58f3c911bd475b716812e7c70ac44614a8","source":{"kind":"arxiv","id":"2607.10371","version":1},"attestation_state":"computed","paper":{"title":"GigaAM Multilingual: Foundation Model for Underrepresented Languages","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"eess.AS","authors_text":"Aleksandr Kutsakov, Alexandr Maximenko, Andrei Kuzmenko, Dmitrii Bolotov, Fyodor Minkin, Georgii Gospodinov, Oleg Kutuzov, Pavel Bogomolov","submitted_at":"2026-07-11T15:48:03Z","abstract_excerpt":"Despite recent scaling successes, multilingual ASR performance remains highly uneven, with long-tail languages suffering from severe data scarcity. This work addresses the challenge of building robust foundation models for underrepresented Central Asian languages (Kazakh, Kyrgyz, Uzbek). We present GigaAM Multilingual, a Conformer encoder pre-trained on 2M hours of audio using a HuBERT-style objective. Crucially, we introduce a cluster-level data balancing strategy during pre-training and a domain-aware sampling method during fine-tuning to mitigate head-language dominance. In controlled compa"},"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":"2607.10371","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2026-07-11T15:48:03Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"048bb8f52c622b0ae68e5e57c14a2938d2da46e6704187d67a915e23e8ee7bad","abstract_canon_sha256":"d0b5e3868064c0893c6ca3c40a4e211fe9b7095ea6e72f3399ad884e1d88e137"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T01:21:12.519783Z","signature_b64":"Ey2mjs5MLzXW1LgoBp2RVVybbUMa8gR7cl5Q9Gq3SxMT2n+eeCg3AfO3pnRDFMamLfrBKSV1WgqmPYkgxRbpDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c92bdf7b4b216997f8495a3bde9b4f58f3c911bd475b716812e7c70ac44614a8","last_reissued_at":"2026-07-14T01:21:12.518901Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T01:21:12.518901Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GigaAM Multilingual: Foundation Model for Underrepresented Languages","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"eess.AS","authors_text":"Aleksandr Kutsakov, Alexandr Maximenko, Andrei Kuzmenko, Dmitrii Bolotov, Fyodor Minkin, Georgii Gospodinov, Oleg Kutuzov, Pavel Bogomolov","submitted_at":"2026-07-11T15:48:03Z","abstract_excerpt":"Despite recent scaling successes, multilingual ASR performance remains highly uneven, with long-tail languages suffering from severe data scarcity. This work addresses the challenge of building robust foundation models for underrepresented Central Asian languages (Kazakh, Kyrgyz, Uzbek). We present GigaAM Multilingual, a Conformer encoder pre-trained on 2M hours of audio using a HuBERT-style objective. Crucially, we introduce a cluster-level data balancing strategy during pre-training and a domain-aware sampling method during fine-tuning to mitigate head-language dominance. In controlled compa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.10371","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/2607.10371/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":"2607.10371","created_at":"2026-07-14T01:21:12.519361+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.10371v1","created_at":"2026-07-14T01:21:12.519361+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.10371","created_at":"2026-07-14T01:21:12.519361+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZEV5662LEFUZ","created_at":"2026-07-14T01:21:12.519361+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZEV5662LEFUZP6CJ","created_at":"2026-07-14T01:21:12.519361+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZEV5662L","created_at":"2026-07-14T01:21:12.519361+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/ZEV5662LEFUZP6CJLI555G2PLD","json":"https://pith.science/pith/ZEV5662LEFUZP6CJLI555G2PLD.json","graph_json":"https://pith.science/api/pith-number/ZEV5662LEFUZP6CJLI555G2PLD/graph.json","events_json":"https://pith.science/api/pith-number/ZEV5662LEFUZP6CJLI555G2PLD/events.json","paper":"https://pith.science/paper/ZEV5662L"},"agent_actions":{"view_html":"https://pith.science/pith/ZEV5662LEFUZP6CJLI555G2PLD","download_json":"https://pith.science/pith/ZEV5662LEFUZP6CJLI555G2PLD.json","view_paper":"https://pith.science/paper/ZEV5662L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.10371&json=true","fetch_graph":"https://pith.science/api/pith-number/ZEV5662LEFUZP6CJLI555G2PLD/graph.json","fetch_events":"https://pith.science/api/pith-number/ZEV5662LEFUZP6CJLI555G2PLD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZEV5662LEFUZP6CJLI555G2PLD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZEV5662LEFUZP6CJLI555G2PLD/action/storage_attestation","attest_author":"https://pith.science/pith/ZEV5662LEFUZP6CJLI555G2PLD/action/author_attestation","sign_citation":"https://pith.science/pith/ZEV5662LEFUZP6CJLI555G2PLD/action/citation_signature","submit_replication":"https://pith.science/pith/ZEV5662LEFUZP6CJLI555G2PLD/action/replication_record"}},"created_at":"2026-07-14T01:21:12.519361+00:00","updated_at":"2026-07-14T01:21:12.519361+00:00"}