{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4JV7LUIANJ4XEETGA5XWZADKB7","short_pith_number":"pith:4JV7LUIA","schema_version":"1.0","canonical_sha256":"e26bf5d1006a79721266076f6c806a0fda905e80e4e5890bf6d6f3f827b0654e","source":{"kind":"arxiv","id":"2607.01617","version":1},"attestation_state":"computed","paper":{"title":"3DLS: A 3D Logic-Stacked Architecture for Disaggregated LLM Serving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"In-Jun Jung, Jaehun Lee, Joo-Young Kim","submitted_at":"2026-07-02T02:32:44Z","abstract_excerpt":"Large language model (LLM) serving increasingly combines prefill-decode (PD) disaggregation with tensor parallelism (TP) to support large models and long contexts. In conventional 2D/2.5D chiplet architectures, layer-wise prefill-to-decode KV-cache transfer decode-side TP collectives share the same lateral die-to-die (D2D) interconnect, creating mixed-traffic contention on the decode critical path. This contention increases communication latency, prolongs token generation intervals, and degrades end-to-end serving performance. We propose 3DLS, a logic-on-logic 3D-stacked chiplet architecture t"},"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":"2607.01617","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2026-07-02T02:32:44Z","cross_cats_sorted":[],"title_canon_sha256":"28cbcd7711457b34da037fd90fd4e97b6946cb970f2b32b685c51cd548d0ef0b","abstract_canon_sha256":"874a55b2796d747a00b913bde5eef7e4289dcdc809fc138b797676904ca5443e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T01:17:03.456396Z","signature_b64":"wZLrDQA4DVTx87528dLRia6bgx2XFnhC5ac5/PsE6idh8jtKBz8aL8S0QQWleF04gdt4jhE9Mn2nasDzc/kvAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e26bf5d1006a79721266076f6c806a0fda905e80e4e5890bf6d6f3f827b0654e","last_reissued_at":"2026-07-03T01:17:03.456014Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T01:17:03.456014Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"3DLS: A 3D Logic-Stacked Architecture for Disaggregated LLM Serving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"In-Jun Jung, Jaehun Lee, Joo-Young Kim","submitted_at":"2026-07-02T02:32:44Z","abstract_excerpt":"Large language model (LLM) serving increasingly combines prefill-decode (PD) disaggregation with tensor parallelism (TP) to support large models and long contexts. In conventional 2D/2.5D chiplet architectures, layer-wise prefill-to-decode KV-cache transfer decode-side TP collectives share the same lateral die-to-die (D2D) interconnect, creating mixed-traffic contention on the decode critical path. This contention increases communication latency, prolongs token generation intervals, and degrades end-to-end serving performance. We propose 3DLS, a logic-on-logic 3D-stacked chiplet architecture t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.01617","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/2607.01617/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":"2607.01617","created_at":"2026-07-03T01:17:03.456076+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.01617v1","created_at":"2026-07-03T01:17:03.456076+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.01617","created_at":"2026-07-03T01:17:03.456076+00:00"},{"alias_kind":"pith_short_12","alias_value":"4JV7LUIANJ4X","created_at":"2026-07-03T01:17:03.456076+00:00"},{"alias_kind":"pith_short_16","alias_value":"4JV7LUIANJ4XEETG","created_at":"2026-07-03T01:17:03.456076+00:00"},{"alias_kind":"pith_short_8","alias_value":"4JV7LUIA","created_at":"2026-07-03T01:17:03.456076+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/4JV7LUIANJ4XEETGA5XWZADKB7","json":"https://pith.science/pith/4JV7LUIANJ4XEETGA5XWZADKB7.json","graph_json":"https://pith.science/api/pith-number/4JV7LUIANJ4XEETGA5XWZADKB7/graph.json","events_json":"https://pith.science/api/pith-number/4JV7LUIANJ4XEETGA5XWZADKB7/events.json","paper":"https://pith.science/paper/4JV7LUIA"},"agent_actions":{"view_html":"https://pith.science/pith/4JV7LUIANJ4XEETGA5XWZADKB7","download_json":"https://pith.science/pith/4JV7LUIANJ4XEETGA5XWZADKB7.json","view_paper":"https://pith.science/paper/4JV7LUIA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.01617&json=true","fetch_graph":"https://pith.science/api/pith-number/4JV7LUIANJ4XEETGA5XWZADKB7/graph.json","fetch_events":"https://pith.science/api/pith-number/4JV7LUIANJ4XEETGA5XWZADKB7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4JV7LUIANJ4XEETGA5XWZADKB7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4JV7LUIANJ4XEETGA5XWZADKB7/action/storage_attestation","attest_author":"https://pith.science/pith/4JV7LUIANJ4XEETGA5XWZADKB7/action/author_attestation","sign_citation":"https://pith.science/pith/4JV7LUIANJ4XEETGA5XWZADKB7/action/citation_signature","submit_replication":"https://pith.science/pith/4JV7LUIANJ4XEETGA5XWZADKB7/action/replication_record"}},"created_at":"2026-07-03T01:17:03.456076+00:00","updated_at":"2026-07-03T01:17:03.456076+00:00"}