{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RFJCGF4BBHVYSO4APM3UE2EVKC","short_pith_number":"pith:RFJCGF4B","schema_version":"1.0","canonical_sha256":"895223178109eb893b807b374268955095ab36771f459baadd0bb8b8bbfd535f","source":{"kind":"arxiv","id":"2411.12692","version":2},"attestation_state":"computed","paper":{"title":"SparseInfer: Training-free Prediction of Activation Sparsity for Fast LLM Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.PF","authors_text":"Hoeseok Yang, Jiho Shin, Youngmin Yi","submitted_at":"2024-11-19T17:59:12Z","abstract_excerpt":"Leveraging sparsity is crucial for optimizing large language model inference. however, modern LLMs employing SiLU as their activation function exhibit minimal activation sparsity. Recent research has proposed replacing SiLU with ReLU to induce significant activation sparsity and showed no downstream task accuracy degradation through fine tuning. However, taking full advantage of it required training a predictor to estimate this sparsity. In this paper, we introduce SparseInfer, a simple, light weight, and training free predictor for activation sparsity of ReLU field LLMs, in which activation s"},"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":"2411.12692","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.PF","submitted_at":"2024-11-19T17:59:12Z","cross_cats_sorted":[],"title_canon_sha256":"9d01e3cd1375ff67cc4188f5f46f563942715923b672360a796a3185a989a895","abstract_canon_sha256":"ae72a1070d587b33677b1185fba7d5ba30cb7a552941b551df1ba4b10cc3d778"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:04:41.019572Z","signature_b64":"czqpZvQ7/SfAfarTMManJKMBV/BNW/8qaL1GuoianuGrdnx0ESAx8q3vNKj9xKBfvpTPeg72ZJHUM6dxPR6FBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"895223178109eb893b807b374268955095ab36771f459baadd0bb8b8bbfd535f","last_reissued_at":"2026-07-05T10:04:41.019167Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:04:41.019167Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SparseInfer: Training-free Prediction of Activation Sparsity for Fast LLM Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.PF","authors_text":"Hoeseok Yang, Jiho Shin, Youngmin Yi","submitted_at":"2024-11-19T17:59:12Z","abstract_excerpt":"Leveraging sparsity is crucial for optimizing large language model inference. however, modern LLMs employing SiLU as their activation function exhibit minimal activation sparsity. Recent research has proposed replacing SiLU with ReLU to induce significant activation sparsity and showed no downstream task accuracy degradation through fine tuning. However, taking full advantage of it required training a predictor to estimate this sparsity. In this paper, we introduce SparseInfer, a simple, light weight, and training free predictor for activation sparsity of ReLU field LLMs, in which activation s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.12692","kind":"arxiv","version":2},"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/2411.12692/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":"2411.12692","created_at":"2026-07-05T10:04:41.019222+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.12692v2","created_at":"2026-07-05T10:04:41.019222+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.12692","created_at":"2026-07-05T10:04:41.019222+00:00"},{"alias_kind":"pith_short_12","alias_value":"RFJCGF4BBHVY","created_at":"2026-07-05T10:04:41.019222+00:00"},{"alias_kind":"pith_short_16","alias_value":"RFJCGF4BBHVYSO4A","created_at":"2026-07-05T10:04:41.019222+00:00"},{"alias_kind":"pith_short_8","alias_value":"RFJCGF4B","created_at":"2026-07-05T10:04:41.019222+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.14179","citing_title":"A Sparsity Predicting Approach for Large Language Models via Activation Pattern Clustering","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RFJCGF4BBHVYSO4APM3UE2EVKC","json":"https://pith.science/pith/RFJCGF4BBHVYSO4APM3UE2EVKC.json","graph_json":"https://pith.science/api/pith-number/RFJCGF4BBHVYSO4APM3UE2EVKC/graph.json","events_json":"https://pith.science/api/pith-number/RFJCGF4BBHVYSO4APM3UE2EVKC/events.json","paper":"https://pith.science/paper/RFJCGF4B"},"agent_actions":{"view_html":"https://pith.science/pith/RFJCGF4BBHVYSO4APM3UE2EVKC","download_json":"https://pith.science/pith/RFJCGF4BBHVYSO4APM3UE2EVKC.json","view_paper":"https://pith.science/paper/RFJCGF4B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.12692&json=true","fetch_graph":"https://pith.science/api/pith-number/RFJCGF4BBHVYSO4APM3UE2EVKC/graph.json","fetch_events":"https://pith.science/api/pith-number/RFJCGF4BBHVYSO4APM3UE2EVKC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RFJCGF4BBHVYSO4APM3UE2EVKC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RFJCGF4BBHVYSO4APM3UE2EVKC/action/storage_attestation","attest_author":"https://pith.science/pith/RFJCGF4BBHVYSO4APM3UE2EVKC/action/author_attestation","sign_citation":"https://pith.science/pith/RFJCGF4BBHVYSO4APM3UE2EVKC/action/citation_signature","submit_replication":"https://pith.science/pith/RFJCGF4BBHVYSO4APM3UE2EVKC/action/replication_record"}},"created_at":"2026-07-05T10:04:41.019222+00:00","updated_at":"2026-07-05T10:04:41.019222+00:00"}