{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:LZND4V2KALD4LNKKZHZQ6QQEOW","short_pith_number":"pith:LZND4V2K","canonical_record":{"source":{"id":"2506.12417","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-06-14T09:21:52Z","cross_cats_sorted":[],"title_canon_sha256":"b67f69ca6182e1f3abe4b5aed324a0bdf75a6462f3c6125c416a2591a029ae77","abstract_canon_sha256":"a2f410ba13c8133404e72a51163d638c04a627175cf97012d9a2d603b7b60f18"},"schema_version":"1.0"},"canonical_sha256":"5e5a3e574a02c7c5b54ac9f30f42047594e9deca80a1451ae0ebde2ea934935a","source":{"kind":"arxiv","id":"2506.12417","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.12417","created_at":"2026-07-05T11:22:42Z"},{"alias_kind":"arxiv_version","alias_value":"2506.12417v2","created_at":"2026-07-05T11:22:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.12417","created_at":"2026-07-05T11:22:42Z"},{"alias_kind":"pith_short_12","alias_value":"LZND4V2KALD4","created_at":"2026-07-05T11:22:42Z"},{"alias_kind":"pith_short_16","alias_value":"LZND4V2KALD4LNKK","created_at":"2026-07-05T11:22:42Z"},{"alias_kind":"pith_short_8","alias_value":"LZND4V2K","created_at":"2026-07-05T11:22:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:LZND4V2KALD4LNKKZHZQ6QQEOW","target":"record","payload":{"canonical_record":{"source":{"id":"2506.12417","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-06-14T09:21:52Z","cross_cats_sorted":[],"title_canon_sha256":"b67f69ca6182e1f3abe4b5aed324a0bdf75a6462f3c6125c416a2591a029ae77","abstract_canon_sha256":"a2f410ba13c8133404e72a51163d638c04a627175cf97012d9a2d603b7b60f18"},"schema_version":"1.0"},"canonical_sha256":"5e5a3e574a02c7c5b54ac9f30f42047594e9deca80a1451ae0ebde2ea934935a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:22:42.063919Z","signature_b64":"PwayYo6L80FbDIvQhA6+/QUbVFgHSAzhanQhShLG/6YHuScuM+gPyHo0zD3jZ+jM9T+M3+worRniCKRRIx6YCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5e5a3e574a02c7c5b54ac9f30f42047594e9deca80a1451ae0ebde2ea934935a","last_reissued_at":"2026-07-05T11:22:42.063378Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:22:42.063378Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.12417","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:22:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TR00BUPjZuSWBMvUBuJf24IvJkkHZoCr1am+qRSgCYtbQxFqDdPB4o/NuSJ+gvZO3MfOeDctYxzSiN00pxcMAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T03:14:01.766789Z"},"content_sha256":"2cbe1f12f9d0680fdb950dea1d9ae98fbf6ff4258812c907aeb3c129ea3ec67c","schema_version":"1.0","event_id":"sha256:2cbe1f12f9d0680fdb950dea1d9ae98fbf6ff4258812c907aeb3c129ea3ec67c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:LZND4V2KALD4LNKKZHZQ6QQEOW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"HarMoEny: Efficient Multi-GPU Inference of MoE Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Anne-Marie Kermarrec, Martijn de Vos, Oana Balmau, Rafael Pires, Rishi Sharma, Zachary Doucet","submitted_at":"2025-06-14T09:21:52Z","abstract_excerpt":"Mixture-of-Experts (MoE) models offer computational efficiency during inference by activating only a subset of specialized experts for a given input. This enables efficient model scaling on multi-GPU systems that use expert parallelism without compromising performance. However, load imbalance among experts and GPUs introduces waiting times, which can significantly increase inference latency. To address this challenge, we propose HarMoEny, a novel solution to address MoE load imbalance through two simple techniques: (i) dynamic token redistribution to underutilized GPUs and (ii) asynchronous pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.12417","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/2506.12417/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":"2026-07-05T11:22:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tzU5KFB+BYrjNSkCzWHPoXYHKv+aofaVnjS0wutpaE1+zp9BxvK4k+fqiUfbkZdpcaPKCyuGve7GKldkvnrWBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T03:14:01.767319Z"},"content_sha256":"503196abcf65cd66f27b032dd30083bd2a699c2c38ddc0334db87c9d568229cd","schema_version":"1.0","event_id":"sha256:503196abcf65cd66f27b032dd30083bd2a699c2c38ddc0334db87c9d568229cd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LZND4V2KALD4LNKKZHZQ6QQEOW/bundle.json","state_url":"https://pith.science/pith/LZND4V2KALD4LNKKZHZQ6QQEOW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LZND4V2KALD4LNKKZHZQ6QQEOW/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":"2026-08-21T03:14:01Z","links":{"resolver":"https://pith.science/pith/LZND4V2KALD4LNKKZHZQ6QQEOW","bundle":"https://pith.science/pith/LZND4V2KALD4LNKKZHZQ6QQEOW/bundle.json","state":"https://pith.science/pith/LZND4V2KALD4LNKKZHZQ6QQEOW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LZND4V2KALD4LNKKZHZQ6QQEOW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:LZND4V2KALD4LNKKZHZQ6QQEOW","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":"a2f410ba13c8133404e72a51163d638c04a627175cf97012d9a2d603b7b60f18","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-06-14T09:21:52Z","title_canon_sha256":"b67f69ca6182e1f3abe4b5aed324a0bdf75a6462f3c6125c416a2591a029ae77"},"schema_version":"1.0","source":{"id":"2506.12417","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.12417","created_at":"2026-07-05T11:22:42Z"},{"alias_kind":"arxiv_version","alias_value":"2506.12417v2","created_at":"2026-07-05T11:22:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.12417","created_at":"2026-07-05T11:22:42Z"},{"alias_kind":"pith_short_12","alias_value":"LZND4V2KALD4","created_at":"2026-07-05T11:22:42Z"},{"alias_kind":"pith_short_16","alias_value":"LZND4V2KALD4LNKK","created_at":"2026-07-05T11:22:42Z"},{"alias_kind":"pith_short_8","alias_value":"LZND4V2K","created_at":"2026-07-05T11:22:42Z"}],"graph_snapshots":[{"event_id":"sha256:503196abcf65cd66f27b032dd30083bd2a699c2c38ddc0334db87c9d568229cd","target":"graph","created_at":"2026-07-05T11:22:42Z","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/2506.12417/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Mixture-of-Experts (MoE) models offer computational efficiency during inference by activating only a subset of specialized experts for a given input. This enables efficient model scaling on multi-GPU systems that use expert parallelism without compromising performance. However, load imbalance among experts and GPUs introduces waiting times, which can significantly increase inference latency. To address this challenge, we propose HarMoEny, a novel solution to address MoE load imbalance through two simple techniques: (i) dynamic token redistribution to underutilized GPUs and (ii) asynchronous pr","authors_text":"Anne-Marie Kermarrec, Martijn de Vos, Oana Balmau, Rafael Pires, Rishi Sharma, Zachary Doucet","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-06-14T09:21:52Z","title":"HarMoEny: Efficient Multi-GPU Inference of MoE Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.12417","kind":"arxiv","version":2},"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":"sha256:2cbe1f12f9d0680fdb950dea1d9ae98fbf6ff4258812c907aeb3c129ea3ec67c","target":"record","created_at":"2026-07-05T11:22:42Z","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":"a2f410ba13c8133404e72a51163d638c04a627175cf97012d9a2d603b7b60f18","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-06-14T09:21:52Z","title_canon_sha256":"b67f69ca6182e1f3abe4b5aed324a0bdf75a6462f3c6125c416a2591a029ae77"},"schema_version":"1.0","source":{"id":"2506.12417","kind":"arxiv","version":2}},"canonical_sha256":"5e5a3e574a02c7c5b54ac9f30f42047594e9deca80a1451ae0ebde2ea934935a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5e5a3e574a02c7c5b54ac9f30f42047594e9deca80a1451ae0ebde2ea934935a","first_computed_at":"2026-07-05T11:22:42.063378Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:22:42.063378Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PwayYo6L80FbDIvQhA6+/QUbVFgHSAzhanQhShLG/6YHuScuM+gPyHo0zD3jZ+jM9T+M3+worRniCKRRIx6YCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:22:42.063919Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.12417","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2cbe1f12f9d0680fdb950dea1d9ae98fbf6ff4258812c907aeb3c129ea3ec67c","sha256:503196abcf65cd66f27b032dd30083bd2a699c2c38ddc0334db87c9d568229cd"],"state_sha256":"149762f0546bc6fdae66079e68805ed27d8c014984b0be3ed890a40633d64710"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vaDFTNkEr7pa8KgxE3jbqpMq4XbpITqv3p5crxxaLexv7J7vfzQwnHrZPslTYA4+KRl8Nl8g/4jnZbTf8Na6BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-21T03:14:01.771617Z","bundle_sha256":"6ed17c4ddc5f5cb4d164c6e0d6242d5c496a1b6f4e667778b7918e960f56afbd"}}