{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:Y44ARQ3A7R37MA3UJEILH3MPVI","short_pith_number":"pith:Y44ARQ3A","canonical_record":{"source":{"id":"2501.06589","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-11T17:06:30Z","cross_cats_sorted":["cs.CL","cs.DC"],"title_canon_sha256":"5c8632ad6e4918920dd9f0671be888426f33eb3e25a3d96ee4a0a4c10b024aea","abstract_canon_sha256":"e99aa1cd8c245f0ea8b4888e3476b82b9ce925ca8218363fbb03e6441197ee21"},"schema_version":"1.0"},"canonical_sha256":"c73808c360fc77f603744910b3ed8faa0396346407aff1d37b969020be0b9b9d","source":{"kind":"arxiv","id":"2501.06589","version":5},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.06589","created_at":"2026-07-05T11:24:12Z"},{"alias_kind":"arxiv_version","alias_value":"2501.06589v5","created_at":"2026-07-05T11:24:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.06589","created_at":"2026-07-05T11:24:12Z"},{"alias_kind":"pith_short_12","alias_value":"Y44ARQ3A7R37","created_at":"2026-07-05T11:24:12Z"},{"alias_kind":"pith_short_16","alias_value":"Y44ARQ3A7R37MA3U","created_at":"2026-07-05T11:24:12Z"},{"alias_kind":"pith_short_8","alias_value":"Y44ARQ3A","created_at":"2026-07-05T11:24:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:Y44ARQ3A7R37MA3UJEILH3MPVI","target":"record","payload":{"canonical_record":{"source":{"id":"2501.06589","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-11T17:06:30Z","cross_cats_sorted":["cs.CL","cs.DC"],"title_canon_sha256":"5c8632ad6e4918920dd9f0671be888426f33eb3e25a3d96ee4a0a4c10b024aea","abstract_canon_sha256":"e99aa1cd8c245f0ea8b4888e3476b82b9ce925ca8218363fbb03e6441197ee21"},"schema_version":"1.0"},"canonical_sha256":"c73808c360fc77f603744910b3ed8faa0396346407aff1d37b969020be0b9b9d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:24:12.550654Z","signature_b64":"pfn8Wh+kJAxaKyuONRcTGmQm2Kq91Y9RQKrAR4qu/kVTloJVviUogk62Zl0xe90YxV+z5kZZALgmw6IKmUrnAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c73808c360fc77f603744910b3ed8faa0396346407aff1d37b969020be0b9b9d","last_reissued_at":"2026-07-05T11:24:12.550086Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:24:12.550086Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.06589","source_version":5,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:24:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HzeiTIO5RwQKPgb0Y+9NDRNmp9drj7oRdVySbIPhCU59hcFpKua4pwL0H3Fj3pVxFTuBn2IuKpMictl/BOssDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T16:59:59.215037Z"},"content_sha256":"f2a5760f2e5d3c587ce271cd38c12ee7699a92989837c17cb7731f740fae7688","schema_version":"1.0","event_id":"sha256:f2a5760f2e5d3c587ce271cd38c12ee7699a92989837c17cb7731f740fae7688"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:Y44ARQ3A7R37MA3UJEILH3MPVI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Ladder-residual: parallelism-aware architecture for accelerating large model inference with communication overlapping","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.DC"],"primary_cat":"cs.LG","authors_text":"Ben Athiwaratkun, Jonathan Ragan-Kelley, Jue Wang, Mayank Mishra, Muru Zhang, Shuaiwen Leon Song, Tri Dao, William Brandon, Yoon Kim, Zhongzhu Zhou","submitted_at":"2025-01-11T17:06:30Z","abstract_excerpt":"Large language model inference is both memory-intensive and time-consuming, often requiring distributed algorithms to efficiently scale. Various model parallelism strategies are used in multi-gpu training and inference to partition computation across multiple devices, reducing memory load and computation time. However, using model parallelism necessitates communication of information between GPUs, which has been a major bottleneck and limits the gains obtained by scaling up the number of devices. We introduce Ladder Residual, a simple architectural modification applicable to all residual-based"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.06589","kind":"arxiv","version":5},"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/2501.06589/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:24:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zufnw/XCTfvHQZVTQlJ4joMZ10JaWhaRSCLt39IbzkFJZYl3/Hw8EKF1jgMUfmvy5qu7LKmuR330CzqOFNRbCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T16:59:59.215547Z"},"content_sha256":"57a02b24863b64d2c1bb0cf3c7c28e7a5749d6b7507ed9c3dabd7d6d487010c1","schema_version":"1.0","event_id":"sha256:57a02b24863b64d2c1bb0cf3c7c28e7a5749d6b7507ed9c3dabd7d6d487010c1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Y44ARQ3A7R37MA3UJEILH3MPVI/bundle.json","state_url":"https://pith.science/pith/Y44ARQ3A7R37MA3UJEILH3MPVI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Y44ARQ3A7R37MA3UJEILH3MPVI/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T16:59:59Z","links":{"resolver":"https://pith.science/pith/Y44ARQ3A7R37MA3UJEILH3MPVI","bundle":"https://pith.science/pith/Y44ARQ3A7R37MA3UJEILH3MPVI/bundle.json","state":"https://pith.science/pith/Y44ARQ3A7R37MA3UJEILH3MPVI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Y44ARQ3A7R37MA3UJEILH3MPVI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:Y44ARQ3A7R37MA3UJEILH3MPVI","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"e99aa1cd8c245f0ea8b4888e3476b82b9ce925ca8218363fbb03e6441197ee21","cross_cats_sorted":["cs.CL","cs.DC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-11T17:06:30Z","title_canon_sha256":"5c8632ad6e4918920dd9f0671be888426f33eb3e25a3d96ee4a0a4c10b024aea"},"schema_version":"1.0","source":{"id":"2501.06589","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.06589","created_at":"2026-07-05T11:24:12Z"},{"alias_kind":"arxiv_version","alias_value":"2501.06589v5","created_at":"2026-07-05T11:24:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.06589","created_at":"2026-07-05T11:24:12Z"},{"alias_kind":"pith_short_12","alias_value":"Y44ARQ3A7R37","created_at":"2026-07-05T11:24:12Z"},{"alias_kind":"pith_short_16","alias_value":"Y44ARQ3A7R37MA3U","created_at":"2026-07-05T11:24:12Z"},{"alias_kind":"pith_short_8","alias_value":"Y44ARQ3A","created_at":"2026-07-05T11:24:12Z"}],"graph_snapshots":[{"event_id":"sha256:57a02b24863b64d2c1bb0cf3c7c28e7a5749d6b7507ed9c3dabd7d6d487010c1","target":"graph","created_at":"2026-07-05T11:24:12Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2501.06589/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language model inference is both memory-intensive and time-consuming, often requiring distributed algorithms to efficiently scale. Various model parallelism strategies are used in multi-gpu training and inference to partition computation across multiple devices, reducing memory load and computation time. However, using model parallelism necessitates communication of information between GPUs, which has been a major bottleneck and limits the gains obtained by scaling up the number of devices. We introduce Ladder Residual, a simple architectural modification applicable to all residual-based","authors_text":"Ben Athiwaratkun, Jonathan Ragan-Kelley, Jue Wang, Mayank Mishra, Muru Zhang, Shuaiwen Leon Song, Tri Dao, William Brandon, Yoon Kim, Zhongzhu Zhou","cross_cats":["cs.CL","cs.DC"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-11T17:06:30Z","title":"Ladder-residual: parallelism-aware architecture for accelerating large model inference with communication overlapping"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.06589","kind":"arxiv","version":5},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f2a5760f2e5d3c587ce271cd38c12ee7699a92989837c17cb7731f740fae7688","target":"record","created_at":"2026-07-05T11:24:12Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"e99aa1cd8c245f0ea8b4888e3476b82b9ce925ca8218363fbb03e6441197ee21","cross_cats_sorted":["cs.CL","cs.DC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-11T17:06:30Z","title_canon_sha256":"5c8632ad6e4918920dd9f0671be888426f33eb3e25a3d96ee4a0a4c10b024aea"},"schema_version":"1.0","source":{"id":"2501.06589","kind":"arxiv","version":5}},"canonical_sha256":"c73808c360fc77f603744910b3ed8faa0396346407aff1d37b969020be0b9b9d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c73808c360fc77f603744910b3ed8faa0396346407aff1d37b969020be0b9b9d","first_computed_at":"2026-07-05T11:24:12.550086Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:24:12.550086Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pfn8Wh+kJAxaKyuONRcTGmQm2Kq91Y9RQKrAR4qu/kVTloJVviUogk62Zl0xe90YxV+z5kZZALgmw6IKmUrnAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:24:12.550654Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.06589","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f2a5760f2e5d3c587ce271cd38c12ee7699a92989837c17cb7731f740fae7688","sha256:57a02b24863b64d2c1bb0cf3c7c28e7a5749d6b7507ed9c3dabd7d6d487010c1"],"state_sha256":"243856b8d945caea1e73d1a0c8ba81833adc279b5e830ecec2678bc3559ed527"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/pkCDMgjMb+kHgO0o9dn8hLrjxRpbMRri+lSgjcKLsIsgS6ExaajTlSiuTaKyft2lEIJ/ReOOy91A7Ywj4QfAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T16:59:59.220294Z","bundle_sha256":"37a84e997f11aa8f17eb6beba93854dd79254699d653fd2c78ba10cc150b4f06"}}