{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:APEUIO7KERA7QI6KETHDBSG7PG","short_pith_number":"pith:APEUIO7K","schema_version":"1.0","canonical_sha256":"03c9443bea2441f823ca24ce30c8df799aa0bf2c6c6c0e9893e7f142c1c48b75","source":{"kind":"arxiv","id":"2509.25214","version":4},"attestation_state":"computed","paper":{"title":"On-the-Fly Adaptation to Quantization: Configuration-Aware LoRA for Efficient Fine-Tuning of Quantized LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A single configuration-aware model generates effective LoRA adjustments for any quantization setting of an LLM without retraining per configuration.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Edith C. H. Ngai, Ming Tang, Rongguang Ye","submitted_at":"2025-09-22T11:07:50Z","abstract_excerpt":"As increasingly large pre-trained models are released, deploying them on edge devices for privacy-preserving applications requires effective compression. Recent works combine quantization with the fine-tuning of high-precision LoRA adapters, which can substantially reduce model size while mitigating the accuracy loss from quantization. However, edge devices have inherently heterogeneous capabilities, while performing configuration-wise fine-tuning for every quantization setting is computationally prohibitive. In this paper, we propose CoA-LoRA, a method that dynamically adjusts the LoRA adapte"},"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":true},"canonical_record":{"source":{"id":"2509.25214","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-09-22T11:07:50Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"abbfc27828e4645993eee55e963cc2ba42e476eabf42094228462acdad3626c7","abstract_canon_sha256":"206675ea5466bbc53114fd921373a1cf2f9c685a2a656bca4697848b74d6ecc9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:17.155681Z","signature_b64":"didfKrpQmYS20G/x7yRH2vGsR/Poh9Gkzcv5Ujr4Vwb0qE0NCtzCZlZCRRWK5x0Qnj74aU4Jk0l73+qo8BWtAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"03c9443bea2441f823ca24ce30c8df799aa0bf2c6c6c0e9893e7f142c1c48b75","last_reissued_at":"2026-06-23T02:13:17.155161Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:17.155161Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On-the-Fly Adaptation to Quantization: Configuration-Aware LoRA for Efficient Fine-Tuning of Quantized LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A single configuration-aware model generates effective LoRA adjustments for any quantization setting of an LLM without retraining per configuration.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Edith C. H. Ngai, Ming Tang, Rongguang Ye","submitted_at":"2025-09-22T11:07:50Z","abstract_excerpt":"As increasingly large pre-trained models are released, deploying them on edge devices for privacy-preserving applications requires effective compression. Recent works combine quantization with the fine-tuning of high-precision LoRA adapters, which can substantially reduce model size while mitigating the accuracy loss from quantization. However, edge devices have inherently heterogeneous capabilities, while performing configuration-wise fine-tuning for every quantization setting is computationally prohibitive. In this paper, we propose CoA-LoRA, a method that dynamically adjusts the LoRA adapte"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"CoA-LoRA dynamically adjusts the LoRA adapter to arbitrary quantization configurations without requiring repeated fine-tuning and achieves comparable or superior performance to state-of-the-art methods that fine-tune a separate LoRA adapter for each configuration.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The configuration-aware model can accurately predict low-rank adjustments for unseen quantization configurations when trained only on a Pareto-selected subset of configurations that cover different total bit-width budgets.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CoA-LoRA trains a single configuration-aware model on a Pareto-optimized set of quantization configurations to enable dynamic LoRA adaptation to arbitrary bit-width assignments without per-configuration fine-tuning.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A single configuration-aware model generates effective LoRA adjustments for any quantization setting of an LLM without retraining per configuration.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"feac72112f5f4f2b596bfb1a9ff5e67f9fa16e4eb4d458372db1fcb8c227d842"},"source":{"id":"2509.25214","kind":"arxiv","version":4},"verdict":{"id":"005bcfd4-597f-4294-9493-a6a2f653299c","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-18T13:49:52.475415Z","strongest_claim":"CoA-LoRA dynamically adjusts the LoRA adapter to arbitrary quantization configurations without requiring repeated fine-tuning and achieves comparable or superior performance to state-of-the-art methods that fine-tune a separate LoRA adapter for each configuration.","one_line_summary":"CoA-LoRA trains a single configuration-aware model on a Pareto-optimized set of quantization configurations to enable dynamic LoRA adaptation to arbitrary bit-width assignments without per-configuration fine-tuning.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The configuration-aware model can accurately predict low-rank adjustments for unseen quantization configurations when trained only on a Pareto-selected subset of configurations that cover different total bit-width budgets.","pith_extraction_headline":"A single configuration-aware model generates effective LoRA adjustments for any quantization setting of an LLM without retraining per configuration."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2509.25214/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":2,"snapshot_sha256":"d5a6c1ab72ef4a06457ff43482d5ad8e5646e8a7a417f930a2881632d3ba2ca1"},"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":"2509.25214","created_at":"2026-06-23T02:13:17.155220+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.25214v4","created_at":"2026-06-23T02:13:17.155220+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.25214","created_at":"2026-06-23T02:13:17.155220+00:00"},{"alias_kind":"pith_short_12","alias_value":"APEUIO7KERA7","created_at":"2026-06-23T02:13:17.155220+00:00"},{"alias_kind":"pith_short_16","alias_value":"APEUIO7KERA7QI6K","created_at":"2026-06-23T02:13:17.155220+00:00"},{"alias_kind":"pith_short_8","alias_value":"APEUIO7K","created_at":"2026-06-23T02:13:17.155220+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.00494","citing_title":"ProjQ: Project-and-Quantize for Adapter-Aware LLM Compression","ref_index":46,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/APEUIO7KERA7QI6KETHDBSG7PG","json":"https://pith.science/pith/APEUIO7KERA7QI6KETHDBSG7PG.json","graph_json":"https://pith.science/api/pith-number/APEUIO7KERA7QI6KETHDBSG7PG/graph.json","events_json":"https://pith.science/api/pith-number/APEUIO7KERA7QI6KETHDBSG7PG/events.json","paper":"https://pith.science/paper/APEUIO7K"},"agent_actions":{"view_html":"https://pith.science/pith/APEUIO7KERA7QI6KETHDBSG7PG","download_json":"https://pith.science/pith/APEUIO7KERA7QI6KETHDBSG7PG.json","view_paper":"https://pith.science/paper/APEUIO7K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.25214&json=true","fetch_graph":"https://pith.science/api/pith-number/APEUIO7KERA7QI6KETHDBSG7PG/graph.json","fetch_events":"https://pith.science/api/pith-number/APEUIO7KERA7QI6KETHDBSG7PG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/APEUIO7KERA7QI6KETHDBSG7PG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/APEUIO7KERA7QI6KETHDBSG7PG/action/storage_attestation","attest_author":"https://pith.science/pith/APEUIO7KERA7QI6KETHDBSG7PG/action/author_attestation","sign_citation":"https://pith.science/pith/APEUIO7KERA7QI6KETHDBSG7PG/action/citation_signature","submit_replication":"https://pith.science/pith/APEUIO7KERA7QI6KETHDBSG7PG/action/replication_record"}},"created_at":"2026-06-23T02:13:17.155220+00:00","updated_at":"2026-06-23T02:13:17.155220+00:00"}