{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2ILPWV45ZCVD3PKUKQJI7URNJR","short_pith_number":"pith:2ILPWV45","schema_version":"1.0","canonical_sha256":"d216fb579dc8aa3dbd5454128fd22d4c5e8c6cfb50515df48698e6965b9eb66f","source":{"kind":"arxiv","id":"2409.11233","version":1},"attestation_state":"computed","paper":{"title":"Evaluating the Impact of Compression Techniques on Task-Specific Performance of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bishwash Khanal, Jeffery M. Capone","submitted_at":"2024-09-17T14:34:11Z","abstract_excerpt":"Large language models (LLMs) offer powerful capabilities but incur substantial computational costs, driving the need for efficient compression techniques. This study evaluates the impact of popular compression methods - Magnitude Pruning, SparseGPT, and Wanda - on the LLaMA-2-7B model, focusing on the trade-offs between model size reduction, downstream task performance, and the role of calibration data. Our findings reveal that while SparseGPT and Wanda preserve perplexity even at 50% sparsity, they suffer significant degradation on downstream tasks, highlighting the inadequacy of perplexity a"},"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":"2409.11233","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-17T14:34:11Z","cross_cats_sorted":[],"title_canon_sha256":"2211d27c36cc26b4b1239ce57386bacc6b7ccebbc243dc519130a352dec65223","abstract_canon_sha256":"da99c5795fe728a0c5c96d7025b5c695c8ee65bbaf593cb5582de8a65a17bf3a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:08:04.074701Z","signature_b64":"keb7mgASKsxaE2zaSRNjmdVIXFulG718ajr4A/rlHY+G1We/KVPN6qe/t6nysliL2gK+0A2GIlwdS6/+2fOcBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d216fb579dc8aa3dbd5454128fd22d4c5e8c6cfb50515df48698e6965b9eb66f","last_reissued_at":"2026-07-05T09:08:04.074263Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:08:04.074263Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluating the Impact of Compression Techniques on Task-Specific Performance of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bishwash Khanal, Jeffery M. Capone","submitted_at":"2024-09-17T14:34:11Z","abstract_excerpt":"Large language models (LLMs) offer powerful capabilities but incur substantial computational costs, driving the need for efficient compression techniques. This study evaluates the impact of popular compression methods - Magnitude Pruning, SparseGPT, and Wanda - on the LLaMA-2-7B model, focusing on the trade-offs between model size reduction, downstream task performance, and the role of calibration data. Our findings reveal that while SparseGPT and Wanda preserve perplexity even at 50% sparsity, they suffer significant degradation on downstream tasks, highlighting the inadequacy of perplexity a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.11233","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/2409.11233/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":"2409.11233","created_at":"2026-07-05T09:08:04.074326+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.11233v1","created_at":"2026-07-05T09:08:04.074326+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.11233","created_at":"2026-07-05T09:08:04.074326+00:00"},{"alias_kind":"pith_short_12","alias_value":"2ILPWV45ZCVD","created_at":"2026-07-05T09:08:04.074326+00:00"},{"alias_kind":"pith_short_16","alias_value":"2ILPWV45ZCVD3PKU","created_at":"2026-07-05T09:08:04.074326+00:00"},{"alias_kind":"pith_short_8","alias_value":"2ILPWV45","created_at":"2026-07-05T09:08:04.074326+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19993","citing_title":"Activation- and Influence-Aware Ranks (AIR): Function-Preserving SVD Compression for LLMs","ref_index":184,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2ILPWV45ZCVD3PKUKQJI7URNJR","json":"https://pith.science/pith/2ILPWV45ZCVD3PKUKQJI7URNJR.json","graph_json":"https://pith.science/api/pith-number/2ILPWV45ZCVD3PKUKQJI7URNJR/graph.json","events_json":"https://pith.science/api/pith-number/2ILPWV45ZCVD3PKUKQJI7URNJR/events.json","paper":"https://pith.science/paper/2ILPWV45"},"agent_actions":{"view_html":"https://pith.science/pith/2ILPWV45ZCVD3PKUKQJI7URNJR","download_json":"https://pith.science/pith/2ILPWV45ZCVD3PKUKQJI7URNJR.json","view_paper":"https://pith.science/paper/2ILPWV45","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.11233&json=true","fetch_graph":"https://pith.science/api/pith-number/2ILPWV45ZCVD3PKUKQJI7URNJR/graph.json","fetch_events":"https://pith.science/api/pith-number/2ILPWV45ZCVD3PKUKQJI7URNJR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2ILPWV45ZCVD3PKUKQJI7URNJR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2ILPWV45ZCVD3PKUKQJI7URNJR/action/storage_attestation","attest_author":"https://pith.science/pith/2ILPWV45ZCVD3PKUKQJI7URNJR/action/author_attestation","sign_citation":"https://pith.science/pith/2ILPWV45ZCVD3PKUKQJI7URNJR/action/citation_signature","submit_replication":"https://pith.science/pith/2ILPWV45ZCVD3PKUKQJI7URNJR/action/replication_record"}},"created_at":"2026-07-05T09:08:04.074326+00:00","updated_at":"2026-07-05T09:08:04.074326+00:00"}