{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QQRSCYTMYBE4ONSOBVGZSLJU4N","short_pith_number":"pith:QQRSCYTM","schema_version":"1.0","canonical_sha256":"842321626cc049c7364e0d4d992d34e362850c8ca8e91a88eb1423e55309eba1","source":{"kind":"arxiv","id":"2408.01415","version":1},"attestation_state":"computed","paper":{"title":"Conditional LoRA Parameter Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Dongwen Tang, Junshu Tang, Kai Wang, Wangbo Zhao, Xiaolong Jin, Yang You, Yukun Zhou","submitted_at":"2024-08-02T17:43:34Z","abstract_excerpt":"Generative models have achieved remarkable success in image, video, and text domains. Inspired by this, researchers have explored utilizing generative models to generate neural network parameters. However, these efforts have been limited by the parameter size and the practicality of generating high-performance parameters. In this paper, we propose COND P-DIFF, a novel approach that demonstrates the feasibility of controllable high-performance parameter generation, particularly for LoRA (Low-Rank Adaptation) weights, during the fine-tuning process. Specifically, we employ an autoencoder to extr"},"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":"2408.01415","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-08-02T17:43:34Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"eedcfa8d0cd4f7e9e0ac400c781f31c233b68eaebf62da4b3c5f3886248e2f82","abstract_canon_sha256":"0d356f9727fee86285ce84177e73b4e7845d4d3f8b9694a21fb2c41ea44e1544"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:51:38.122104Z","signature_b64":"E0khAqSLfhWQeiYiUrcI156G8YJ+ETVNRFGVRRGaYb4VjW8wAwNVTu/DGnOymIYFgtyhFH2n28tVFgGRmELhBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"842321626cc049c7364e0d4d992d34e362850c8ca8e91a88eb1423e55309eba1","last_reissued_at":"2026-07-05T08:51:38.121639Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:51:38.121639Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Conditional LoRA Parameter Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Dongwen Tang, Junshu Tang, Kai Wang, Wangbo Zhao, Xiaolong Jin, Yang You, Yukun Zhou","submitted_at":"2024-08-02T17:43:34Z","abstract_excerpt":"Generative models have achieved remarkable success in image, video, and text domains. Inspired by this, researchers have explored utilizing generative models to generate neural network parameters. However, these efforts have been limited by the parameter size and the practicality of generating high-performance parameters. In this paper, we propose COND P-DIFF, a novel approach that demonstrates the feasibility of controllable high-performance parameter generation, particularly for LoRA (Low-Rank Adaptation) weights, during the fine-tuning process. Specifically, we employ an autoencoder to extr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.01415","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/2408.01415/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":"2408.01415","created_at":"2026-07-05T08:51:38.121693+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.01415v1","created_at":"2026-07-05T08:51:38.121693+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.01415","created_at":"2026-07-05T08:51:38.121693+00:00"},{"alias_kind":"pith_short_12","alias_value":"QQRSCYTMYBE4","created_at":"2026-07-05T08:51:38.121693+00:00"},{"alias_kind":"pith_short_16","alias_value":"QQRSCYTMYBE4ONSO","created_at":"2026-07-05T08:51:38.121693+00:00"},{"alias_kind":"pith_short_8","alias_value":"QQRSCYTM","created_at":"2026-07-05T08:51:38.121693+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26772","citing_title":"Escaping Iterative Parameter-Space Noise: Differentially Private Learning with a Hypernetwork","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07217","citing_title":"Robotic Policy Adaptation via Weight-Space Meta-Learning","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02606","citing_title":"ReLoRA: Knowledge-Reusing Adaptation for Fast Rollout of Evolving LLM Services","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20189","citing_title":"SOLAR: A Self-Optimizing Open-Ended Autonomous Agent for Lifelong Learning and Continual Adaptation","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13839","citing_title":"Good Agentic Friends Do Not Just Give Verbal Advice: They Can Update Your Weights","ref_index":42,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QQRSCYTMYBE4ONSOBVGZSLJU4N","json":"https://pith.science/pith/QQRSCYTMYBE4ONSOBVGZSLJU4N.json","graph_json":"https://pith.science/api/pith-number/QQRSCYTMYBE4ONSOBVGZSLJU4N/graph.json","events_json":"https://pith.science/api/pith-number/QQRSCYTMYBE4ONSOBVGZSLJU4N/events.json","paper":"https://pith.science/paper/QQRSCYTM"},"agent_actions":{"view_html":"https://pith.science/pith/QQRSCYTMYBE4ONSOBVGZSLJU4N","download_json":"https://pith.science/pith/QQRSCYTMYBE4ONSOBVGZSLJU4N.json","view_paper":"https://pith.science/paper/QQRSCYTM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.01415&json=true","fetch_graph":"https://pith.science/api/pith-number/QQRSCYTMYBE4ONSOBVGZSLJU4N/graph.json","fetch_events":"https://pith.science/api/pith-number/QQRSCYTMYBE4ONSOBVGZSLJU4N/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QQRSCYTMYBE4ONSOBVGZSLJU4N/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QQRSCYTMYBE4ONSOBVGZSLJU4N/action/storage_attestation","attest_author":"https://pith.science/pith/QQRSCYTMYBE4ONSOBVGZSLJU4N/action/author_attestation","sign_citation":"https://pith.science/pith/QQRSCYTMYBE4ONSOBVGZSLJU4N/action/citation_signature","submit_replication":"https://pith.science/pith/QQRSCYTMYBE4ONSOBVGZSLJU4N/action/replication_record"}},"created_at":"2026-07-05T08:51:38.121693+00:00","updated_at":"2026-07-05T08:51:38.121693+00:00"}