{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IUCAJPPXPEQFYOXHKOUYTFHEYQ","short_pith_number":"pith:IUCAJPPX","schema_version":"1.0","canonical_sha256":"450404bdf779205c3ae753a98994e4c415cdcc765236a231aa421dd8a7e1ad5e","source":{"kind":"arxiv","id":"2502.00899","version":1},"attestation_state":"computed","paper":{"title":"HASSLE-free: A unified Framework for Sparse plus Low-Rank Matrix Decomposition for LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Kayhan Behdin, Mehdi Makni, Natalia Ponomareva, Rahul Mazumder, Zheng Xu","submitted_at":"2025-02-02T20:23:32Z","abstract_excerpt":"The impressive capabilities of large foundation models come at a cost of substantial computing resources to serve them. Compressing these pre-trained models is of practical interest as it can democratize deploying them to the machine learning community at large by lowering the costs associated with inference. A promising compression scheme is to decompose foundation models' dense weights into a sum of sparse plus low-rank matrices. In this paper, we design a unified framework coined HASSLE-free for (semi-structured) sparse plus low-rank matrix decomposition of foundation models. Our framework "},"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":"2502.00899","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2025-02-02T20:23:32Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2ec75294f38b46cf749a3f8712f7415fc26f9bc72c2ecbf648cd1a22a6b3c45e","abstract_canon_sha256":"919136d9be35b987db7d5a7ae0aa388d77902a677bb0b24439a3347322bd0d14"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:08:39.854441Z","signature_b64":"NRp0Ga3tBD6QhjbxFAzlbLVSTEm9C/xAHZnk4/06PsHwVjmP37f+ODvbajllqMyvtK64H3HuDjesZ/50deW9BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"450404bdf779205c3ae753a98994e4c415cdcc765236a231aa421dd8a7e1ad5e","last_reissued_at":"2026-07-05T10:08:39.853945Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:08:39.853945Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HASSLE-free: A unified Framework for Sparse plus Low-Rank Matrix Decomposition for LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Kayhan Behdin, Mehdi Makni, Natalia Ponomareva, Rahul Mazumder, Zheng Xu","submitted_at":"2025-02-02T20:23:32Z","abstract_excerpt":"The impressive capabilities of large foundation models come at a cost of substantial computing resources to serve them. Compressing these pre-trained models is of practical interest as it can democratize deploying them to the machine learning community at large by lowering the costs associated with inference. A promising compression scheme is to decompose foundation models' dense weights into a sum of sparse plus low-rank matrices. In this paper, we design a unified framework coined HASSLE-free for (semi-structured) sparse plus low-rank matrix decomposition of foundation models. Our framework "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.00899","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/2502.00899/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":"2502.00899","created_at":"2026-07-05T10:08:39.854003+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.00899v1","created_at":"2026-07-05T10:08:39.854003+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.00899","created_at":"2026-07-05T10:08:39.854003+00:00"},{"alias_kind":"pith_short_12","alias_value":"IUCAJPPXPEQF","created_at":"2026-07-05T10:08:39.854003+00:00"},{"alias_kind":"pith_short_16","alias_value":"IUCAJPPXPEQFYOXH","created_at":"2026-07-05T10:08:39.854003+00:00"},{"alias_kind":"pith_short_8","alias_value":"IUCAJPPX","created_at":"2026-07-05T10:08:39.854003+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03465","citing_title":"Rethinking the Role of Tensor Decompositions in Post-Training LLM Compression","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03667","citing_title":"ELAS: Efficient Pre-Training of Low-Rank Large Language Models via 2:4 Activation Sparsity","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IUCAJPPXPEQFYOXHKOUYTFHEYQ","json":"https://pith.science/pith/IUCAJPPXPEQFYOXHKOUYTFHEYQ.json","graph_json":"https://pith.science/api/pith-number/IUCAJPPXPEQFYOXHKOUYTFHEYQ/graph.json","events_json":"https://pith.science/api/pith-number/IUCAJPPXPEQFYOXHKOUYTFHEYQ/events.json","paper":"https://pith.science/paper/IUCAJPPX"},"agent_actions":{"view_html":"https://pith.science/pith/IUCAJPPXPEQFYOXHKOUYTFHEYQ","download_json":"https://pith.science/pith/IUCAJPPXPEQFYOXHKOUYTFHEYQ.json","view_paper":"https://pith.science/paper/IUCAJPPX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.00899&json=true","fetch_graph":"https://pith.science/api/pith-number/IUCAJPPXPEQFYOXHKOUYTFHEYQ/graph.json","fetch_events":"https://pith.science/api/pith-number/IUCAJPPXPEQFYOXHKOUYTFHEYQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IUCAJPPXPEQFYOXHKOUYTFHEYQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IUCAJPPXPEQFYOXHKOUYTFHEYQ/action/storage_attestation","attest_author":"https://pith.science/pith/IUCAJPPXPEQFYOXHKOUYTFHEYQ/action/author_attestation","sign_citation":"https://pith.science/pith/IUCAJPPXPEQFYOXHKOUYTFHEYQ/action/citation_signature","submit_replication":"https://pith.science/pith/IUCAJPPXPEQFYOXHKOUYTFHEYQ/action/replication_record"}},"created_at":"2026-07-05T10:08:39.854003+00:00","updated_at":"2026-07-05T10:08:39.854003+00:00"}