{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:AGJZ3TWDX2QWFUT4NBCCNTXYFS","short_pith_number":"pith:AGJZ3TWD","schema_version":"1.0","canonical_sha256":"01939dcec3bea162d27c684426cef82cb84b65f2f3a215a1db781f170916c4fd","source":{"kind":"arxiv","id":"2501.16337","version":1},"attestation_state":"computed","paper":{"title":"Explore Activation Sparsity in Recurrent LLMs for Energy-Efficient Neuromorphic Computing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.AR","cs.LG"],"primary_cat":"cs.NE","authors_text":"Amirreza Yousefzadeh, Guangzhi Tang, Ivan Knunyants, Manolis Sifalakis, Maryam Tavakol, YingFu Xu","submitted_at":"2025-01-09T19:13:03Z","abstract_excerpt":"The recent rise of Large Language Models (LLMs) has revolutionized the deep learning field. However, the desire to deploy LLMs on edge devices introduces energy efficiency and latency challenges. Recurrent LLM (R-LLM) architectures have proven effective in mitigating the quadratic complexity of self-attention, making them a potential paradigm for computing on-edge neuromorphic processors. In this work, we propose a low-cost, training-free algorithm to sparsify R-LLMs' activations to enhance energy efficiency on neuromorphic hardware. Our approach capitalizes on the inherent structure of these "},"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":"2501.16337","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.NE","submitted_at":"2025-01-09T19:13:03Z","cross_cats_sorted":["cs.AI","cs.AR","cs.LG"],"title_canon_sha256":"912690e4f1a844f61b0b039d7948bd3a8e9a167eeb42d2f3398a8b142397ea02","abstract_canon_sha256":"94c392b278e7a54f7aefcb54766121b489d6da5a18f55cd56ccef0db20e45fcd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:06:11.537988Z","signature_b64":"FiFuEEbZn9fhkyyIGdJaNMvGiiuA7xqa37DSOREQFQehBXYR4/EbcRB2ZQczb+grSPPEuTWgt7se8NenOJbwBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"01939dcec3bea162d27c684426cef82cb84b65f2f3a215a1db781f170916c4fd","last_reissued_at":"2026-07-05T10:06:11.537441Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:06:11.537441Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Explore Activation Sparsity in Recurrent LLMs for Energy-Efficient Neuromorphic Computing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.AR","cs.LG"],"primary_cat":"cs.NE","authors_text":"Amirreza Yousefzadeh, Guangzhi Tang, Ivan Knunyants, Manolis Sifalakis, Maryam Tavakol, YingFu Xu","submitted_at":"2025-01-09T19:13:03Z","abstract_excerpt":"The recent rise of Large Language Models (LLMs) has revolutionized the deep learning field. However, the desire to deploy LLMs on edge devices introduces energy efficiency and latency challenges. Recurrent LLM (R-LLM) architectures have proven effective in mitigating the quadratic complexity of self-attention, making them a potential paradigm for computing on-edge neuromorphic processors. In this work, we propose a low-cost, training-free algorithm to sparsify R-LLMs' activations to enhance energy efficiency on neuromorphic hardware. Our approach capitalizes on the inherent structure of these "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.16337","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/2501.16337/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":"2501.16337","created_at":"2026-07-05T10:06:11.537508+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.16337v1","created_at":"2026-07-05T10:06:11.537508+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.16337","created_at":"2026-07-05T10:06:11.537508+00:00"},{"alias_kind":"pith_short_12","alias_value":"AGJZ3TWDX2QW","created_at":"2026-07-05T10:06:11.537508+00:00"},{"alias_kind":"pith_short_16","alias_value":"AGJZ3TWDX2QWFUT4","created_at":"2026-07-05T10:06:11.537508+00:00"},{"alias_kind":"pith_short_8","alias_value":"AGJZ3TWD","created_at":"2026-07-05T10:06:11.537508+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/AGJZ3TWDX2QWFUT4NBCCNTXYFS","json":"https://pith.science/pith/AGJZ3TWDX2QWFUT4NBCCNTXYFS.json","graph_json":"https://pith.science/api/pith-number/AGJZ3TWDX2QWFUT4NBCCNTXYFS/graph.json","events_json":"https://pith.science/api/pith-number/AGJZ3TWDX2QWFUT4NBCCNTXYFS/events.json","paper":"https://pith.science/paper/AGJZ3TWD"},"agent_actions":{"view_html":"https://pith.science/pith/AGJZ3TWDX2QWFUT4NBCCNTXYFS","download_json":"https://pith.science/pith/AGJZ3TWDX2QWFUT4NBCCNTXYFS.json","view_paper":"https://pith.science/paper/AGJZ3TWD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.16337&json=true","fetch_graph":"https://pith.science/api/pith-number/AGJZ3TWDX2QWFUT4NBCCNTXYFS/graph.json","fetch_events":"https://pith.science/api/pith-number/AGJZ3TWDX2QWFUT4NBCCNTXYFS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AGJZ3TWDX2QWFUT4NBCCNTXYFS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AGJZ3TWDX2QWFUT4NBCCNTXYFS/action/storage_attestation","attest_author":"https://pith.science/pith/AGJZ3TWDX2QWFUT4NBCCNTXYFS/action/author_attestation","sign_citation":"https://pith.science/pith/AGJZ3TWDX2QWFUT4NBCCNTXYFS/action/citation_signature","submit_replication":"https://pith.science/pith/AGJZ3TWDX2QWFUT4NBCCNTXYFS/action/replication_record"}},"created_at":"2026-07-05T10:06:11.537508+00:00","updated_at":"2026-07-05T10:06:11.537508+00:00"}