{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:RGKUW6Z5YHP6PPLU6A64TC277C","short_pith_number":"pith:RGKUW6Z5","schema_version":"1.0","canonical_sha256":"89954b7b3dc1dfe7bd74f03dc98b5ff8be5a64702031f3bd56013dfbea24c28c","source":{"kind":"arxiv","id":"2606.22968","version":1},"attestation_state":"computed","paper":{"title":"MOCAP: Wafer-Scale-Chip-Oriented Memory-Orchestrated Chunked Pipelining Framework for Prefill-Only LLM Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Chao Li, Haonan Zuo, Huizheng Wang, Jinyi Deng, Shouyi Yin, Taiquan Wei, Yang Hu, Yuheng Xiao, Zichuan Wang","submitted_at":"2026-06-22T07:51:15Z","abstract_excerpt":"Large language models (LLMs) are increasingly used in prefill-only workloads, where end-to-end latency is dominated by the prefill phase. For long-context prefill, communication overhead grows with sequence length and quickly becomes a bottleneck on conventional GPU systems, making wafer-scale chips (WSCs) a promising substrate due to their high communication bandwidth and large aggregate compute and memory capacity. A natural way to accelerate prefill is to partition a long input sequence into multiple chunks and execute them in a finer-grained pipeline across devices. However, directly apply"},"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":"2606.22968","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2026-06-22T07:51:15Z","cross_cats_sorted":[],"title_canon_sha256":"1820ffce405dc5a3628feb6e915bfd155e80dccf3f0b091f444540f45844bbf8","abstract_canon_sha256":"ce333928222271fbe9d8012133488816b0601ffb9648d771ed767cb7e2548b20"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T03:14:05.516245Z","signature_b64":"8U4+ehfP/bg+SKwiLpi8zm3Uw11kwzcsSriTm6pynl0LhKvkAE44zQa14Wh67nCcev6vcAwBxJL10GJFfdd7BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"89954b7b3dc1dfe7bd74f03dc98b5ff8be5a64702031f3bd56013dfbea24c28c","last_reissued_at":"2026-06-23T03:14:05.515868Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T03:14:05.515868Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MOCAP: Wafer-Scale-Chip-Oriented Memory-Orchestrated Chunked Pipelining Framework for Prefill-Only LLM Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Chao Li, Haonan Zuo, Huizheng Wang, Jinyi Deng, Shouyi Yin, Taiquan Wei, Yang Hu, Yuheng Xiao, Zichuan Wang","submitted_at":"2026-06-22T07:51:15Z","abstract_excerpt":"Large language models (LLMs) are increasingly used in prefill-only workloads, where end-to-end latency is dominated by the prefill phase. For long-context prefill, communication overhead grows with sequence length and quickly becomes a bottleneck on conventional GPU systems, making wafer-scale chips (WSCs) a promising substrate due to their high communication bandwidth and large aggregate compute and memory capacity. A natural way to accelerate prefill is to partition a long input sequence into multiple chunks and execute them in a finer-grained pipeline across devices. However, directly apply"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22968","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/2606.22968/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":"2606.22968","created_at":"2026-06-23T03:14:05.515925+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.22968v1","created_at":"2026-06-23T03:14:05.515925+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22968","created_at":"2026-06-23T03:14:05.515925+00:00"},{"alias_kind":"pith_short_12","alias_value":"RGKUW6Z5YHP6","created_at":"2026-06-23T03:14:05.515925+00:00"},{"alias_kind":"pith_short_16","alias_value":"RGKUW6Z5YHP6PPLU","created_at":"2026-06-23T03:14:05.515925+00:00"},{"alias_kind":"pith_short_8","alias_value":"RGKUW6Z5","created_at":"2026-06-23T03:14:05.515925+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/RGKUW6Z5YHP6PPLU6A64TC277C","json":"https://pith.science/pith/RGKUW6Z5YHP6PPLU6A64TC277C.json","graph_json":"https://pith.science/api/pith-number/RGKUW6Z5YHP6PPLU6A64TC277C/graph.json","events_json":"https://pith.science/api/pith-number/RGKUW6Z5YHP6PPLU6A64TC277C/events.json","paper":"https://pith.science/paper/RGKUW6Z5"},"agent_actions":{"view_html":"https://pith.science/pith/RGKUW6Z5YHP6PPLU6A64TC277C","download_json":"https://pith.science/pith/RGKUW6Z5YHP6PPLU6A64TC277C.json","view_paper":"https://pith.science/paper/RGKUW6Z5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.22968&json=true","fetch_graph":"https://pith.science/api/pith-number/RGKUW6Z5YHP6PPLU6A64TC277C/graph.json","fetch_events":"https://pith.science/api/pith-number/RGKUW6Z5YHP6PPLU6A64TC277C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RGKUW6Z5YHP6PPLU6A64TC277C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RGKUW6Z5YHP6PPLU6A64TC277C/action/storage_attestation","attest_author":"https://pith.science/pith/RGKUW6Z5YHP6PPLU6A64TC277C/action/author_attestation","sign_citation":"https://pith.science/pith/RGKUW6Z5YHP6PPLU6A64TC277C/action/citation_signature","submit_replication":"https://pith.science/pith/RGKUW6Z5YHP6PPLU6A64TC277C/action/replication_record"}},"created_at":"2026-06-23T03:14:05.515925+00:00","updated_at":"2026-06-23T03:14:05.515925+00:00"}