{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VBKJNNPNDAKROCX2JYGIJRICHH","short_pith_number":"pith:VBKJNNPN","schema_version":"1.0","canonical_sha256":"a85496b5ed1815170afa4e0c84c50239d9f9c1c57d57980724dcf9ce61cea7f8","source":{"kind":"arxiv","id":"2412.19925","version":2},"attestation_state":"computed","paper":{"title":"HADES: Hardware Accelerated Decoding for Efficient Speculation in Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.AR"],"primary_cat":"cs.CL","authors_text":"Xinhe Xu, Yihong Jin, Ze Yang","submitted_at":"2024-12-27T21:19:01Z","abstract_excerpt":"Large Language Models (LLMs) have revolutionized natural language processing by understanding and generating human-like text. However, the increasing demand for more sophisticated LLMs presents significant computational challenges due to their scale and complexity. This paper introduces Hardware Accelerated Decoding (HADES), a novel approach to enhance the performance and energy efficiency of LLMs. We address the design of an LLM accelerator with hardware-level speculative decoding support, a concept not previously explored in existing literature. Our work demonstrates how speculative decoding"},"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":"2412.19925","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-27T21:19:01Z","cross_cats_sorted":["cs.AI","cs.AR"],"title_canon_sha256":"15d5c3f393db2c52c686aa04130a477405d51862afa741272941a31a877d99f1","abstract_canon_sha256":"c3cc2172cb80b2c2a656a9c3dc2792b243ec207c59504c3da3055f6738358f20"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:59:57.993212Z","signature_b64":"AQ/qMvedb8NMzVzNtdIQAcUJYAdCqyhNUjMLn+IiTjfQJ8I8LNQQT4S0UNz0wBKUyzG9a946Hc7/Xsi8YVGwCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a85496b5ed1815170afa4e0c84c50239d9f9c1c57d57980724dcf9ce61cea7f8","last_reissued_at":"2026-07-05T09:59:57.992791Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:59:57.992791Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HADES: Hardware Accelerated Decoding for Efficient Speculation in Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.AR"],"primary_cat":"cs.CL","authors_text":"Xinhe Xu, Yihong Jin, Ze Yang","submitted_at":"2024-12-27T21:19:01Z","abstract_excerpt":"Large Language Models (LLMs) have revolutionized natural language processing by understanding and generating human-like text. However, the increasing demand for more sophisticated LLMs presents significant computational challenges due to their scale and complexity. This paper introduces Hardware Accelerated Decoding (HADES), a novel approach to enhance the performance and energy efficiency of LLMs. We address the design of an LLM accelerator with hardware-level speculative decoding support, a concept not previously explored in existing literature. Our work demonstrates how speculative decoding"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.19925","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/2412.19925/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":"2412.19925","created_at":"2026-07-05T09:59:57.992848+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.19925v2","created_at":"2026-07-05T09:59:57.992848+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.19925","created_at":"2026-07-05T09:59:57.992848+00:00"},{"alias_kind":"pith_short_12","alias_value":"VBKJNNPNDAKR","created_at":"2026-07-05T09:59:57.992848+00:00"},{"alias_kind":"pith_short_16","alias_value":"VBKJNNPNDAKROCX2","created_at":"2026-07-05T09:59:57.992848+00:00"},{"alias_kind":"pith_short_8","alias_value":"VBKJNNPN","created_at":"2026-07-05T09:59:57.992848+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01595","citing_title":"Safe and Adaptive Cloud Healing: Verifying LLM-Generated Recovery Plans with a Neural-Symbolic World Model","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VBKJNNPNDAKROCX2JYGIJRICHH","json":"https://pith.science/pith/VBKJNNPNDAKROCX2JYGIJRICHH.json","graph_json":"https://pith.science/api/pith-number/VBKJNNPNDAKROCX2JYGIJRICHH/graph.json","events_json":"https://pith.science/api/pith-number/VBKJNNPNDAKROCX2JYGIJRICHH/events.json","paper":"https://pith.science/paper/VBKJNNPN"},"agent_actions":{"view_html":"https://pith.science/pith/VBKJNNPNDAKROCX2JYGIJRICHH","download_json":"https://pith.science/pith/VBKJNNPNDAKROCX2JYGIJRICHH.json","view_paper":"https://pith.science/paper/VBKJNNPN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.19925&json=true","fetch_graph":"https://pith.science/api/pith-number/VBKJNNPNDAKROCX2JYGIJRICHH/graph.json","fetch_events":"https://pith.science/api/pith-number/VBKJNNPNDAKROCX2JYGIJRICHH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VBKJNNPNDAKROCX2JYGIJRICHH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VBKJNNPNDAKROCX2JYGIJRICHH/action/storage_attestation","attest_author":"https://pith.science/pith/VBKJNNPNDAKROCX2JYGIJRICHH/action/author_attestation","sign_citation":"https://pith.science/pith/VBKJNNPNDAKROCX2JYGIJRICHH/action/citation_signature","submit_replication":"https://pith.science/pith/VBKJNNPNDAKROCX2JYGIJRICHH/action/replication_record"}},"created_at":"2026-07-05T09:59:57.992848+00:00","updated_at":"2026-07-05T09:59:57.992848+00:00"}