{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:RPDBSNAMR6HEQ3EAKIAIMVWOXR","short_pith_number":"pith:RPDBSNAM","schema_version":"1.0","canonical_sha256":"8bc619340c8f8e486c8052008656cebc68429c3ac069866c6a63e4d2a8bf0f73","source":{"kind":"arxiv","id":"2606.08994","version":1},"attestation_state":"computed","paper":{"title":"Language-Aware Token Boosting: LLM Language Confusion Reduction Without Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Nut Chukamphaeng, Pakhapoom Sarapat, Trapoom Ukarapol","submitted_at":"2026-06-08T03:47:07Z","abstract_excerpt":"Large language models (LLMs) sometimes exhibit language confusion when generating non-English text. Existing approaches typically rely on fine-tuning to mitigate this issue. In contrast, we propose a tuning-free paradigm for reducing language confusion. Within this paradigm, we introduce two methods: Language-Aware Token Boosting (LATB), which applies targeted perturbations to tokens associated with the desired language, and Adaptive Language-Aware Token Boosting (Adaptive-LATB), which dynamically adjusts these perturbations based on the model's confidence in the intended language. Experiments"},"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":"2606.08994","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-08T03:47:07Z","cross_cats_sorted":[],"title_canon_sha256":"4ee06f7468c0b78a71712e7ecdde9213b28a0b286a9ed934aedd3e5b22865c80","abstract_canon_sha256":"4eb5f27c8bba3b347a0d83cfa5ef6cd0f0cd20a13b75fb9b429ad2050e1f7723"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T02:07:52.522285Z","signature_b64":"a6YcsbA2NNbJ6hBaVQSOudfaHwhWNY/JRuT+TWEMdezHQwYXl8u/MpGPOnFqDYm9rzQ4Vl9e3emnyTZENLh+Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8bc619340c8f8e486c8052008656cebc68429c3ac069866c6a63e4d2a8bf0f73","last_reissued_at":"2026-06-09T02:07:52.521310Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T02:07:52.521310Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Language-Aware Token Boosting: LLM Language Confusion Reduction Without Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Nut Chukamphaeng, Pakhapoom Sarapat, Trapoom Ukarapol","submitted_at":"2026-06-08T03:47:07Z","abstract_excerpt":"Large language models (LLMs) sometimes exhibit language confusion when generating non-English text. Existing approaches typically rely on fine-tuning to mitigate this issue. In contrast, we propose a tuning-free paradigm for reducing language confusion. Within this paradigm, we introduce two methods: Language-Aware Token Boosting (LATB), which applies targeted perturbations to tokens associated with the desired language, and Adaptive Language-Aware Token Boosting (Adaptive-LATB), which dynamically adjusts these perturbations based on the model's confidence in the intended language. Experiments"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.08994","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/2606.08994/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":"2606.08994","created_at":"2026-06-09T02:07:52.521475+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.08994v1","created_at":"2026-06-09T02:07:52.521475+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.08994","created_at":"2026-06-09T02:07:52.521475+00:00"},{"alias_kind":"pith_short_12","alias_value":"RPDBSNAMR6HE","created_at":"2026-06-09T02:07:52.521475+00:00"},{"alias_kind":"pith_short_16","alias_value":"RPDBSNAMR6HEQ3EA","created_at":"2026-06-09T02:07:52.521475+00:00"},{"alias_kind":"pith_short_8","alias_value":"RPDBSNAM","created_at":"2026-06-09T02:07:52.521475+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/RPDBSNAMR6HEQ3EAKIAIMVWOXR","json":"https://pith.science/pith/RPDBSNAMR6HEQ3EAKIAIMVWOXR.json","graph_json":"https://pith.science/api/pith-number/RPDBSNAMR6HEQ3EAKIAIMVWOXR/graph.json","events_json":"https://pith.science/api/pith-number/RPDBSNAMR6HEQ3EAKIAIMVWOXR/events.json","paper":"https://pith.science/paper/RPDBSNAM"},"agent_actions":{"view_html":"https://pith.science/pith/RPDBSNAMR6HEQ3EAKIAIMVWOXR","download_json":"https://pith.science/pith/RPDBSNAMR6HEQ3EAKIAIMVWOXR.json","view_paper":"https://pith.science/paper/RPDBSNAM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.08994&json=true","fetch_graph":"https://pith.science/api/pith-number/RPDBSNAMR6HEQ3EAKIAIMVWOXR/graph.json","fetch_events":"https://pith.science/api/pith-number/RPDBSNAMR6HEQ3EAKIAIMVWOXR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RPDBSNAMR6HEQ3EAKIAIMVWOXR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RPDBSNAMR6HEQ3EAKIAIMVWOXR/action/storage_attestation","attest_author":"https://pith.science/pith/RPDBSNAMR6HEQ3EAKIAIMVWOXR/action/author_attestation","sign_citation":"https://pith.science/pith/RPDBSNAMR6HEQ3EAKIAIMVWOXR/action/citation_signature","submit_replication":"https://pith.science/pith/RPDBSNAMR6HEQ3EAKIAIMVWOXR/action/replication_record"}},"created_at":"2026-06-09T02:07:52.521475+00:00","updated_at":"2026-06-09T02:07:52.521475+00:00"}