{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SA4MT6RFXJ47CQCFFROBBF5V7M","short_pith_number":"pith:SA4MT6RF","schema_version":"1.0","canonical_sha256":"9038c9fa25ba79f140452c5c1097b5fb0a24271b463a03bea83855bd6a137af8","source":{"kind":"arxiv","id":"2405.13816","version":2},"attestation_state":"computed","paper":{"title":"Getting More from Less: Large Language Models are Good Spontaneous Multilingual Learners","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Changjiang Gao, Chao Deng, Jiajun Chen, Junlan Feng, Shimao Zhang, Shujian Huang, Wenhao Zhu, Xin Huang, Xue Han","submitted_at":"2024-05-22T16:46:19Z","abstract_excerpt":"Recently, Large Language Models (LLMs) have shown impressive language capabilities. While most of the existing LLMs have very unbalanced performance across different languages, multilingual alignment based on translation parallel data is an effective method to enhance the LLMs' multilingual capabilities. In this work, we discover and comprehensively investigate the spontaneous multilingual alignment improvement of LLMs. We find that LLMs instruction-tuned on the question translation data (i.e. without annotated answers) are able to encourage the alignment between English and a wide range of la"},"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":"2405.13816","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-22T16:46:19Z","cross_cats_sorted":[],"title_canon_sha256":"d74338aedcafcd868c249894c919cf177e3de30e832f774cc43287e3a9bd2c5b","abstract_canon_sha256":"7155d776e912986b144fb215a3f7c14f9020428a6e1572cc23c847bfc0327ca2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:33:40.074657Z","signature_b64":"l7n3W0TzVD2UHOge3c91b/psODm6CV8OUDhzk1yGdddSZ8L+zK2k7hu2weWbr/IKW5P5VIzm3wIisZL4Z3i3BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9038c9fa25ba79f140452c5c1097b5fb0a24271b463a03bea83855bd6a137af8","last_reissued_at":"2026-07-05T08:33:40.074142Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:33:40.074142Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Getting More from Less: Large Language Models are Good Spontaneous Multilingual Learners","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Changjiang Gao, Chao Deng, Jiajun Chen, Junlan Feng, Shimao Zhang, Shujian Huang, Wenhao Zhu, Xin Huang, Xue Han","submitted_at":"2024-05-22T16:46:19Z","abstract_excerpt":"Recently, Large Language Models (LLMs) have shown impressive language capabilities. While most of the existing LLMs have very unbalanced performance across different languages, multilingual alignment based on translation parallel data is an effective method to enhance the LLMs' multilingual capabilities. In this work, we discover and comprehensively investigate the spontaneous multilingual alignment improvement of LLMs. We find that LLMs instruction-tuned on the question translation data (i.e. without annotated answers) are able to encourage the alignment between English and a wide range of la"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.13816","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/2405.13816/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":"2405.13816","created_at":"2026-07-05T08:33:40.074206+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.13816v2","created_at":"2026-07-05T08:33:40.074206+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.13816","created_at":"2026-07-05T08:33:40.074206+00:00"},{"alias_kind":"pith_short_12","alias_value":"SA4MT6RFXJ47","created_at":"2026-07-05T08:33:40.074206+00:00"},{"alias_kind":"pith_short_16","alias_value":"SA4MT6RFXJ47CQCF","created_at":"2026-07-05T08:33:40.074206+00:00"},{"alias_kind":"pith_short_8","alias_value":"SA4MT6RF","created_at":"2026-07-05T08:33:40.074206+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/SA4MT6RFXJ47CQCFFROBBF5V7M","json":"https://pith.science/pith/SA4MT6RFXJ47CQCFFROBBF5V7M.json","graph_json":"https://pith.science/api/pith-number/SA4MT6RFXJ47CQCFFROBBF5V7M/graph.json","events_json":"https://pith.science/api/pith-number/SA4MT6RFXJ47CQCFFROBBF5V7M/events.json","paper":"https://pith.science/paper/SA4MT6RF"},"agent_actions":{"view_html":"https://pith.science/pith/SA4MT6RFXJ47CQCFFROBBF5V7M","download_json":"https://pith.science/pith/SA4MT6RFXJ47CQCFFROBBF5V7M.json","view_paper":"https://pith.science/paper/SA4MT6RF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.13816&json=true","fetch_graph":"https://pith.science/api/pith-number/SA4MT6RFXJ47CQCFFROBBF5V7M/graph.json","fetch_events":"https://pith.science/api/pith-number/SA4MT6RFXJ47CQCFFROBBF5V7M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SA4MT6RFXJ47CQCFFROBBF5V7M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SA4MT6RFXJ47CQCFFROBBF5V7M/action/storage_attestation","attest_author":"https://pith.science/pith/SA4MT6RFXJ47CQCFFROBBF5V7M/action/author_attestation","sign_citation":"https://pith.science/pith/SA4MT6RFXJ47CQCFFROBBF5V7M/action/citation_signature","submit_replication":"https://pith.science/pith/SA4MT6RFXJ47CQCFFROBBF5V7M/action/replication_record"}},"created_at":"2026-07-05T08:33:40.074206+00:00","updated_at":"2026-07-05T08:33:40.074206+00:00"}