{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:G3RBGRQBLFIHS7PQIJG7YD4WJ3","short_pith_number":"pith:G3RBGRQB","schema_version":"1.0","canonical_sha256":"36e21346015950797df0424dfc0f964eefc7222dceffdd28990a85ffb665bee7","source":{"kind":"arxiv","id":"2606.06985","version":1},"attestation_state":"computed","paper":{"title":"Contrastive Training with LLM-generated Near-Misses for Robust Code-Switching Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Dung D. Le, Giang-Son Nguyen, Hieu Minh Truong, Nhu Vo, Tung X. Nguyen, Wray Buntine","submitted_at":"2026-06-05T07:24:39Z","abstract_excerpt":"Code-switching (CS), the alternation between multiple languages within a single utterance, remains challenging for Automatic Speech Recognition (ASR). To address this issue, we propose a Point-of-Interest (POI)-aware contrastive training framework that improves recognition at CS-critical regions. We first identify CS spans by adopting POI detection method from literature, then construct acoustically plausible near-miss hypotheses by perturbing POIs in ASR N-best outputs and expanding candidates with a large language model. Hard but plausible negatives are retained through filtering with acoust"},"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":"2606.06985","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-05T07:24:39Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"0b0bcc4741eee67d50273349cd1d84238a8fdf572ab9d91a37581120a8ee0f92","abstract_canon_sha256":"6ae35ef8d0caa20037b6c01303254a4f8fe6bb24ba83326a4420ea611b733a2d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-08T01:04:39.618157Z","signature_b64":"FZAShmUaQxsuvU4EqcW0f3O2j/wvASIIfPY7tQPvBq7PBTEeijlEpV+97jgTHDWyoNdjUdCSaT47V6xwT/ouCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"36e21346015950797df0424dfc0f964eefc7222dceffdd28990a85ffb665bee7","last_reissued_at":"2026-06-08T01:04:39.617172Z","signature_status":"signed_v1","first_computed_at":"2026-06-08T01:04:39.617172Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Contrastive Training with LLM-generated Near-Misses for Robust Code-Switching Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Dung D. Le, Giang-Son Nguyen, Hieu Minh Truong, Nhu Vo, Tung X. Nguyen, Wray Buntine","submitted_at":"2026-06-05T07:24:39Z","abstract_excerpt":"Code-switching (CS), the alternation between multiple languages within a single utterance, remains challenging for Automatic Speech Recognition (ASR). To address this issue, we propose a Point-of-Interest (POI)-aware contrastive training framework that improves recognition at CS-critical regions. We first identify CS spans by adopting POI detection method from literature, then construct acoustically plausible near-miss hypotheses by perturbing POIs in ASR N-best outputs and expanding candidates with a large language model. Hard but plausible negatives are retained through filtering with acoust"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.06985","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/2606.06985/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":"2606.06985","created_at":"2026-06-08T01:04:39.617349+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.06985v1","created_at":"2026-06-08T01:04:39.617349+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.06985","created_at":"2026-06-08T01:04:39.617349+00:00"},{"alias_kind":"pith_short_12","alias_value":"G3RBGRQBLFIH","created_at":"2026-06-08T01:04:39.617349+00:00"},{"alias_kind":"pith_short_16","alias_value":"G3RBGRQBLFIHS7PQ","created_at":"2026-06-08T01:04:39.617349+00:00"},{"alias_kind":"pith_short_8","alias_value":"G3RBGRQB","created_at":"2026-06-08T01:04:39.617349+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.06985","citing_title":"Contrastive Training with LLM-generated Near-Misses for Robust Code-Switching Speech Recognition","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G3RBGRQBLFIHS7PQIJG7YD4WJ3","json":"https://pith.science/pith/G3RBGRQBLFIHS7PQIJG7YD4WJ3.json","graph_json":"https://pith.science/api/pith-number/G3RBGRQBLFIHS7PQIJG7YD4WJ3/graph.json","events_json":"https://pith.science/api/pith-number/G3RBGRQBLFIHS7PQIJG7YD4WJ3/events.json","paper":"https://pith.science/paper/G3RBGRQB"},"agent_actions":{"view_html":"https://pith.science/pith/G3RBGRQBLFIHS7PQIJG7YD4WJ3","download_json":"https://pith.science/pith/G3RBGRQBLFIHS7PQIJG7YD4WJ3.json","view_paper":"https://pith.science/paper/G3RBGRQB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.06985&json=true","fetch_graph":"https://pith.science/api/pith-number/G3RBGRQBLFIHS7PQIJG7YD4WJ3/graph.json","fetch_events":"https://pith.science/api/pith-number/G3RBGRQBLFIHS7PQIJG7YD4WJ3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G3RBGRQBLFIHS7PQIJG7YD4WJ3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G3RBGRQBLFIHS7PQIJG7YD4WJ3/action/storage_attestation","attest_author":"https://pith.science/pith/G3RBGRQBLFIHS7PQIJG7YD4WJ3/action/author_attestation","sign_citation":"https://pith.science/pith/G3RBGRQBLFIHS7PQIJG7YD4WJ3/action/citation_signature","submit_replication":"https://pith.science/pith/G3RBGRQBLFIHS7PQIJG7YD4WJ3/action/replication_record"}},"created_at":"2026-06-08T01:04:39.617349+00:00","updated_at":"2026-06-08T01:04:39.617349+00:00"}