{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:LMS2ESP7KZZF6YFBFHHPJEB3T4","short_pith_number":"pith:LMS2ESP7","schema_version":"1.0","canonical_sha256":"5b25a249ff56725f60a129cef4903b9f26c7f28327ea8952da3ebab3cbdbb22a","source":{"kind":"arxiv","id":"2505.24241","version":1},"attestation_state":"computed","paper":{"title":"Advantageous Parameter Expansion Training Makes Better Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"HaiFeng Wang, Hua Wu, Naibin Gu, Peng Fu, Shuohuan Wang, Weiping Wang, Yilong Chen, Yu Sun, Zheng Lin, Zhenyu Zhang","submitted_at":"2025-05-30T06:06:23Z","abstract_excerpt":"Although scaling up the number of trainable parameters in both pre-training and fine-tuning can effectively improve the performance of large language models, it also leads to increased computational overhead. When delving into the parameter difference, we find that a subset of parameters, termed advantageous parameters, plays a crucial role in determining model performance. Further analysis reveals that stronger models tend to possess more such parameters. In this paper, we propose Advantageous Parameter EXpansion Training (APEX), a method that progressively expands advantageous parameters int"},"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":"2505.24241","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T06:06:23Z","cross_cats_sorted":[],"title_canon_sha256":"f182b7b78cb9ccae4dbdd875bc67631567d6a00856c2a621cbb000a3d001e8b2","abstract_canon_sha256":"358a1eff2ca21dc8f98f909db6cd5781fae1f7495b205d6b9d697fa4349b4c60"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:44.862893Z","signature_b64":"xlUalaLkbl0+1QrPK4ilJk6JCz4IqRw3gT5sljY5em6rfHH7lrcUMksczHoF6vVbnXVaQnZctzyU52yDunqVBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5b25a249ff56725f60a129cef4903b9f26c7f28327ea8952da3ebab3cbdbb22a","last_reissued_at":"2026-07-05T11:12:44.862323Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:44.862323Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Advantageous Parameter Expansion Training Makes Better Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"HaiFeng Wang, Hua Wu, Naibin Gu, Peng Fu, Shuohuan Wang, Weiping Wang, Yilong Chen, Yu Sun, Zheng Lin, Zhenyu Zhang","submitted_at":"2025-05-30T06:06:23Z","abstract_excerpt":"Although scaling up the number of trainable parameters in both pre-training and fine-tuning can effectively improve the performance of large language models, it also leads to increased computational overhead. When delving into the parameter difference, we find that a subset of parameters, termed advantageous parameters, plays a crucial role in determining model performance. Further analysis reveals that stronger models tend to possess more such parameters. In this paper, we propose Advantageous Parameter EXpansion Training (APEX), a method that progressively expands advantageous parameters int"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24241","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/2505.24241/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":"2505.24241","created_at":"2026-07-05T11:12:44.862404+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.24241v1","created_at":"2026-07-05T11:12:44.862404+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24241","created_at":"2026-07-05T11:12:44.862404+00:00"},{"alias_kind":"pith_short_12","alias_value":"LMS2ESP7KZZF","created_at":"2026-07-05T11:12:44.862404+00:00"},{"alias_kind":"pith_short_16","alias_value":"LMS2ESP7KZZF6YFB","created_at":"2026-07-05T11:12:44.862404+00:00"},{"alias_kind":"pith_short_8","alias_value":"LMS2ESP7","created_at":"2026-07-05T11:12:44.862404+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/LMS2ESP7KZZF6YFBFHHPJEB3T4","json":"https://pith.science/pith/LMS2ESP7KZZF6YFBFHHPJEB3T4.json","graph_json":"https://pith.science/api/pith-number/LMS2ESP7KZZF6YFBFHHPJEB3T4/graph.json","events_json":"https://pith.science/api/pith-number/LMS2ESP7KZZF6YFBFHHPJEB3T4/events.json","paper":"https://pith.science/paper/LMS2ESP7"},"agent_actions":{"view_html":"https://pith.science/pith/LMS2ESP7KZZF6YFBFHHPJEB3T4","download_json":"https://pith.science/pith/LMS2ESP7KZZF6YFBFHHPJEB3T4.json","view_paper":"https://pith.science/paper/LMS2ESP7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.24241&json=true","fetch_graph":"https://pith.science/api/pith-number/LMS2ESP7KZZF6YFBFHHPJEB3T4/graph.json","fetch_events":"https://pith.science/api/pith-number/LMS2ESP7KZZF6YFBFHHPJEB3T4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LMS2ESP7KZZF6YFBFHHPJEB3T4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LMS2ESP7KZZF6YFBFHHPJEB3T4/action/storage_attestation","attest_author":"https://pith.science/pith/LMS2ESP7KZZF6YFBFHHPJEB3T4/action/author_attestation","sign_citation":"https://pith.science/pith/LMS2ESP7KZZF6YFBFHHPJEB3T4/action/citation_signature","submit_replication":"https://pith.science/pith/LMS2ESP7KZZF6YFBFHHPJEB3T4/action/replication_record"}},"created_at":"2026-07-05T11:12:44.862404+00:00","updated_at":"2026-07-05T11:12:44.862404+00:00"}