{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HHCWXMORGAJUWBLAPMDTC3HBNV","short_pith_number":"pith:HHCWXMOR","schema_version":"1.0","canonical_sha256":"39c56bb1d130134b05607b07316ce16d44ca5c6a04da6227f7e48f3e2d388fdd","source":{"kind":"arxiv","id":"2410.09016","version":3},"attestation_state":"computed","paper":{"title":"Parameter-Efficient Fine-Tuning of State Space Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Hyung Il Koo, Kangwook Lee, Kevin Galim, Wonjun Kang, Yuchen Zeng","submitted_at":"2024-10-11T17:30:28Z","abstract_excerpt":"Deep State Space Models (SSMs), such as Mamba (Gu & Dao, 2024), have become powerful tools for language modeling, offering high performance and linear scalability with sequence length. However, the application of parameter-efficient fine-tuning (PEFT) methods to SSM-based models remains largely underexplored. We start by investigating two fundamental questions on existing PEFT methods: (i) How do they perform on SSM-based models? (ii) Which parameters should they target for optimal results? Our analysis shows that LoRA and its variants consistently outperform all other PEFT methods. While LoRA"},"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":"2410.09016","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-11T17:30:28Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"936c8de1f3d8e912ebe4c9cd3b5b54927e14bb3638d502160c430374ecd2a411","abstract_canon_sha256":"c389b5718bb298075c8ccaa2a9045a63d094c998bc38441d80311c7b5223e795"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:14.800162Z","signature_b64":"TqRwdqUgjsF3wgYWHqWHqJBKhEqCF5D+uR0eBiEMW7Qv4lOvXToaYHK++ksdMCwctTXUf2T6yVHKSZr6Th28Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"39c56bb1d130134b05607b07316ce16d44ca5c6a04da6227f7e48f3e2d388fdd","last_reissued_at":"2026-07-05T11:18:14.799451Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:14.799451Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Parameter-Efficient Fine-Tuning of State Space Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Hyung Il Koo, Kangwook Lee, Kevin Galim, Wonjun Kang, Yuchen Zeng","submitted_at":"2024-10-11T17:30:28Z","abstract_excerpt":"Deep State Space Models (SSMs), such as Mamba (Gu & Dao, 2024), have become powerful tools for language modeling, offering high performance and linear scalability with sequence length. However, the application of parameter-efficient fine-tuning (PEFT) methods to SSM-based models remains largely underexplored. We start by investigating two fundamental questions on existing PEFT methods: (i) How do they perform on SSM-based models? (ii) Which parameters should they target for optimal results? Our analysis shows that LoRA and its variants consistently outperform all other PEFT methods. While LoRA"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.09016","kind":"arxiv","version":3},"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/2410.09016/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":"2410.09016","created_at":"2026-07-05T11:18:14.799556+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.09016v3","created_at":"2026-07-05T11:18:14.799556+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.09016","created_at":"2026-07-05T11:18:14.799556+00:00"},{"alias_kind":"pith_short_12","alias_value":"HHCWXMORGAJU","created_at":"2026-07-05T11:18:14.799556+00:00"},{"alias_kind":"pith_short_16","alias_value":"HHCWXMORGAJUWBLA","created_at":"2026-07-05T11:18:14.799556+00:00"},{"alias_kind":"pith_short_8","alias_value":"HHCWXMOR","created_at":"2026-07-05T11:18:14.799556+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.12770","citing_title":"WriteSAE: Sparse Autoencoders for Recurrent State","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12770","citing_title":"WriteSAE: Sparse Autoencoders for Recurrent State","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12770","citing_title":"WriteSAE: Sparse Autoencoders for Recurrent State","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12770","citing_title":"WriteSAE: Sparse Autoencoders for Recurrent State","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2604.01168","citing_title":"S0 Tuning: Zero-Overhead Adaptation of Hybrid Recurrent-Attention Models","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22127","citing_title":"Where Should LoRA Go? Component-Type Placement in Hybrid Language Models","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HHCWXMORGAJUWBLAPMDTC3HBNV","json":"https://pith.science/pith/HHCWXMORGAJUWBLAPMDTC3HBNV.json","graph_json":"https://pith.science/api/pith-number/HHCWXMORGAJUWBLAPMDTC3HBNV/graph.json","events_json":"https://pith.science/api/pith-number/HHCWXMORGAJUWBLAPMDTC3HBNV/events.json","paper":"https://pith.science/paper/HHCWXMOR"},"agent_actions":{"view_html":"https://pith.science/pith/HHCWXMORGAJUWBLAPMDTC3HBNV","download_json":"https://pith.science/pith/HHCWXMORGAJUWBLAPMDTC3HBNV.json","view_paper":"https://pith.science/paper/HHCWXMOR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.09016&json=true","fetch_graph":"https://pith.science/api/pith-number/HHCWXMORGAJUWBLAPMDTC3HBNV/graph.json","fetch_events":"https://pith.science/api/pith-number/HHCWXMORGAJUWBLAPMDTC3HBNV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HHCWXMORGAJUWBLAPMDTC3HBNV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HHCWXMORGAJUWBLAPMDTC3HBNV/action/storage_attestation","attest_author":"https://pith.science/pith/HHCWXMORGAJUWBLAPMDTC3HBNV/action/author_attestation","sign_citation":"https://pith.science/pith/HHCWXMORGAJUWBLAPMDTC3HBNV/action/citation_signature","submit_replication":"https://pith.science/pith/HHCWXMORGAJUWBLAPMDTC3HBNV/action/replication_record"}},"created_at":"2026-07-05T11:18:14.799556+00:00","updated_at":"2026-07-05T11:18:14.799556+00:00"}