{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4TY4B2OJLMUTRNWHXGGYASOESX","short_pith_number":"pith:4TY4B2OJ","schema_version":"1.0","canonical_sha256":"e4f1c0e9c95b2938b6c7b98d8049c495f3a8b90a411d004bc72a82eef2852955","source":{"kind":"arxiv","id":"2509.00996","version":2},"attestation_state":"computed","paper":{"title":"MEPT: Mixture of Expert Prompt Tuning as a Manifold Mapper","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Cheng Han, Dongfang Liu, Guangyan Sun, Lifu Huang, Qifan Wang, Raghuveer Rao, Runjia Zeng, Sohail Dianat, Tong Geng, Xiaotian Han, Xueling Zhang","submitted_at":"2025-08-31T21:19:25Z","abstract_excerpt":"Considering deep neural networks as manifold mappers, the pretrain-then-fine-tune paradigm can be interpreted as a two-stage process: pretrain establishes a broad knowledge base, and fine-tune adjusts the model parameters to activate specific neural pathways to align with the target manifold. Although prior fine-tuning approaches demonstrate success, their rigid parameter space limits their ability to dynamically activate appropriate neural pathways, rendering them ill-equipped to adapt flexibly to the diverse and evolving data distributions. In light of this view, we propose a novel approach,"},"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":"2509.00996","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-08-31T21:19:25Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"ac93f4422db88fd5ee96defb2d700696ab0ce2a4ce24c220113bec2a08fb61e2","abstract_canon_sha256":"a7a8a6d2f8423dd2e466920ae60989f225f77c8f076206252a72181b62fcaf73"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:11:20.371560Z","signature_b64":"z7U7NrAJ1xcWBC0EEw6VGNgYdSqzDdsjvovMH3hCkVTv2M88SULtx/RH6X6tB9oyd2sfHCXI8cEJRR+YKBUdBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4f1c0e9c95b2938b6c7b98d8049c495f3a8b90a411d004bc72a82eef2852955","last_reissued_at":"2026-07-05T12:11:20.371026Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:11:20.371026Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MEPT: Mixture of Expert Prompt Tuning as a Manifold Mapper","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Cheng Han, Dongfang Liu, Guangyan Sun, Lifu Huang, Qifan Wang, Raghuveer Rao, Runjia Zeng, Sohail Dianat, Tong Geng, Xiaotian Han, Xueling Zhang","submitted_at":"2025-08-31T21:19:25Z","abstract_excerpt":"Considering deep neural networks as manifold mappers, the pretrain-then-fine-tune paradigm can be interpreted as a two-stage process: pretrain establishes a broad knowledge base, and fine-tune adjusts the model parameters to activate specific neural pathways to align with the target manifold. Although prior fine-tuning approaches demonstrate success, their rigid parameter space limits their ability to dynamically activate appropriate neural pathways, rendering them ill-equipped to adapt flexibly to the diverse and evolving data distributions. In light of this view, we propose a novel approach,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.00996","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/2509.00996/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":"2509.00996","created_at":"2026-07-05T12:11:20.371093+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.00996v2","created_at":"2026-07-05T12:11:20.371093+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.00996","created_at":"2026-07-05T12:11:20.371093+00:00"},{"alias_kind":"pith_short_12","alias_value":"4TY4B2OJLMUT","created_at":"2026-07-05T12:11:20.371093+00:00"},{"alias_kind":"pith_short_16","alias_value":"4TY4B2OJLMUTRNWH","created_at":"2026-07-05T12:11:20.371093+00:00"},{"alias_kind":"pith_short_8","alias_value":"4TY4B2OJ","created_at":"2026-07-05T12:11:20.371093+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/4TY4B2OJLMUTRNWHXGGYASOESX","json":"https://pith.science/pith/4TY4B2OJLMUTRNWHXGGYASOESX.json","graph_json":"https://pith.science/api/pith-number/4TY4B2OJLMUTRNWHXGGYASOESX/graph.json","events_json":"https://pith.science/api/pith-number/4TY4B2OJLMUTRNWHXGGYASOESX/events.json","paper":"https://pith.science/paper/4TY4B2OJ"},"agent_actions":{"view_html":"https://pith.science/pith/4TY4B2OJLMUTRNWHXGGYASOESX","download_json":"https://pith.science/pith/4TY4B2OJLMUTRNWHXGGYASOESX.json","view_paper":"https://pith.science/paper/4TY4B2OJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.00996&json=true","fetch_graph":"https://pith.science/api/pith-number/4TY4B2OJLMUTRNWHXGGYASOESX/graph.json","fetch_events":"https://pith.science/api/pith-number/4TY4B2OJLMUTRNWHXGGYASOESX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4TY4B2OJLMUTRNWHXGGYASOESX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4TY4B2OJLMUTRNWHXGGYASOESX/action/storage_attestation","attest_author":"https://pith.science/pith/4TY4B2OJLMUTRNWHXGGYASOESX/action/author_attestation","sign_citation":"https://pith.science/pith/4TY4B2OJLMUTRNWHXGGYASOESX/action/citation_signature","submit_replication":"https://pith.science/pith/4TY4B2OJLMUTRNWHXGGYASOESX/action/replication_record"}},"created_at":"2026-07-05T12:11:20.371093+00:00","updated_at":"2026-07-05T12:11:20.371093+00:00"}