{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SV3ILQS5U6Y4J7DQVAEZXKIUWM","short_pith_number":"pith:SV3ILQS5","schema_version":"1.0","canonical_sha256":"957685c25da7b1c4fc70a8099ba914b30ccbfd4af8befbbc4eac5e115637c272","source":{"kind":"arxiv","id":"2606.00494","version":1},"attestation_state":"computed","paper":{"title":"ProjQ: Project-and-Quantize for Adapter-Aware LLM Compression","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chao Zhang, Li Wang, Merouane Debbah, Samson Lasaulce, Wneya Yu","submitted_at":"2026-05-30T02:54:40Z","abstract_excerpt":"Post-Training Quantization (PTQ) and Low-Rank Adaptation (LoRA) constitute the standard pipeline for efficient Large Language Model (LLM) deployment. However, applying them sequentially poses a problem: PTQ often leaves behind random noise that is spread out (across the model's weights) in a way LoRA can't easily fix, meaning that LoRA ends up wasting its limited capacity trying to fix uncorrectable noise instead of improving task performance. In this paper, we propose \\textbf{ProjQ}, a novel framework for constraining quantization noise to the low-rank manifold via orthogonal subspace project"},"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":"2606.00494","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-30T02:54:40Z","cross_cats_sorted":[],"title_canon_sha256":"c3611788bd450839ec6699010576bee035893f2321d32c0152fee5cc348d614e","abstract_canon_sha256":"a1c5597ec34a703f94cf72b63219b53256e4752c4088f22aabc7da9f87182630"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:56.227752Z","signature_b64":"KVe5YoosVJRa/nGogTpyBSY42nnm8DJR3evmollrV/9wcX1ny6QgeswlMYi1VFn7RETaBw8Sf1f+AykfHyfADA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"957685c25da7b1c4fc70a8099ba914b30ccbfd4af8befbbc4eac5e115637c272","last_reissued_at":"2026-06-02T01:03:56.227364Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:56.227364Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ProjQ: Project-and-Quantize for Adapter-Aware LLM Compression","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chao Zhang, Li Wang, Merouane Debbah, Samson Lasaulce, Wneya Yu","submitted_at":"2026-05-30T02:54:40Z","abstract_excerpt":"Post-Training Quantization (PTQ) and Low-Rank Adaptation (LoRA) constitute the standard pipeline for efficient Large Language Model (LLM) deployment. However, applying them sequentially poses a problem: PTQ often leaves behind random noise that is spread out (across the model's weights) in a way LoRA can't easily fix, meaning that LoRA ends up wasting its limited capacity trying to fix uncorrectable noise instead of improving task performance. In this paper, we propose \\textbf{ProjQ}, a novel framework for constraining quantization noise to the low-rank manifold via orthogonal subspace project"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00494","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/2606.00494/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":"2606.00494","created_at":"2026-06-02T01:03:56.227415+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.00494v1","created_at":"2026-06-02T01:03:56.227415+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00494","created_at":"2026-06-02T01:03:56.227415+00:00"},{"alias_kind":"pith_short_12","alias_value":"SV3ILQS5U6Y4","created_at":"2026-06-02T01:03:56.227415+00:00"},{"alias_kind":"pith_short_16","alias_value":"SV3ILQS5U6Y4J7DQ","created_at":"2026-06-02T01:03:56.227415+00:00"},{"alias_kind":"pith_short_8","alias_value":"SV3ILQS5","created_at":"2026-06-02T01:03:56.227415+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/SV3ILQS5U6Y4J7DQVAEZXKIUWM","json":"https://pith.science/pith/SV3ILQS5U6Y4J7DQVAEZXKIUWM.json","graph_json":"https://pith.science/api/pith-number/SV3ILQS5U6Y4J7DQVAEZXKIUWM/graph.json","events_json":"https://pith.science/api/pith-number/SV3ILQS5U6Y4J7DQVAEZXKIUWM/events.json","paper":"https://pith.science/paper/SV3ILQS5"},"agent_actions":{"view_html":"https://pith.science/pith/SV3ILQS5U6Y4J7DQVAEZXKIUWM","download_json":"https://pith.science/pith/SV3ILQS5U6Y4J7DQVAEZXKIUWM.json","view_paper":"https://pith.science/paper/SV3ILQS5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.00494&json=true","fetch_graph":"https://pith.science/api/pith-number/SV3ILQS5U6Y4J7DQVAEZXKIUWM/graph.json","fetch_events":"https://pith.science/api/pith-number/SV3ILQS5U6Y4J7DQVAEZXKIUWM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SV3ILQS5U6Y4J7DQVAEZXKIUWM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SV3ILQS5U6Y4J7DQVAEZXKIUWM/action/storage_attestation","attest_author":"https://pith.science/pith/SV3ILQS5U6Y4J7DQVAEZXKIUWM/action/author_attestation","sign_citation":"https://pith.science/pith/SV3ILQS5U6Y4J7DQVAEZXKIUWM/action/citation_signature","submit_replication":"https://pith.science/pith/SV3ILQS5U6Y4J7DQVAEZXKIUWM/action/replication_record"}},"created_at":"2026-06-02T01:03:56.227415+00:00","updated_at":"2026-06-02T01:03:56.227415+00:00"}