{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:DUOLM5GFGMSDVWOBAPQLHNWK3F","short_pith_number":"pith:DUOLM5GF","canonical_record":{"source":{"id":"2607.16488","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AR","submitted_at":"2026-07-17T20:27:26Z","cross_cats_sorted":[],"title_canon_sha256":"2274af5f55cdeaf5fdc1880ffaf11f59d36cb238da47159c1ac25f98a53125b4","abstract_canon_sha256":"b4cb20d7130d8fdd2167855c54712651c5986be341f7d0c5485fded1c7e630f9"},"schema_version":"1.0"},"canonical_sha256":"1d1cb674c533243ad9c103e0b3b6cad9515c2cf7a26cfb25a0e8b417d936e248","source":{"kind":"arxiv","id":"2607.16488","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.16488","created_at":"2026-07-21T00:20:21Z"},{"alias_kind":"arxiv_version","alias_value":"2607.16488v1","created_at":"2026-07-21T00:20:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.16488","created_at":"2026-07-21T00:20:21Z"},{"alias_kind":"pith_short_12","alias_value":"DUOLM5GFGMSD","created_at":"2026-07-21T00:20:21Z"},{"alias_kind":"pith_short_16","alias_value":"DUOLM5GFGMSDVWOB","created_at":"2026-07-21T00:20:21Z"},{"alias_kind":"pith_short_8","alias_value":"DUOLM5GF","created_at":"2026-07-21T00:20:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:DUOLM5GFGMSDVWOBAPQLHNWK3F","target":"record","payload":{"canonical_record":{"source":{"id":"2607.16488","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AR","submitted_at":"2026-07-17T20:27:26Z","cross_cats_sorted":[],"title_canon_sha256":"2274af5f55cdeaf5fdc1880ffaf11f59d36cb238da47159c1ac25f98a53125b4","abstract_canon_sha256":"b4cb20d7130d8fdd2167855c54712651c5986be341f7d0c5485fded1c7e630f9"},"schema_version":"1.0"},"canonical_sha256":"1d1cb674c533243ad9c103e0b3b6cad9515c2cf7a26cfb25a0e8b417d936e248","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T00:20:21.217392Z","signature_b64":"HDDMQaTZZVkbknXiFjZYgCCnEjZersbN05GsKM9mh7rxrMfAirkbje/JwUE80hEJevOb20JHBvYX53iG/+C5Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1d1cb674c533243ad9c103e0b3b6cad9515c2cf7a26cfb25a0e8b417d936e248","last_reissued_at":"2026-07-21T00:20:21.216478Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T00:20:21.216478Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.16488","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-21T00:20:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"n7QYxkg41csjAYo1sOJeGsCpHr5TeJcTIXAGll9HW+4YZ7p0F5fednGxiDXOqf0N1saIHxM19TuGIJMGXoauCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T21:20:58.532591Z"},"content_sha256":"f758722ef21b0f388510d002a477ce701d39aaf851f01e4f125e943d8631a342","schema_version":"1.0","event_id":"sha256:f758722ef21b0f388510d002a477ce701d39aaf851f01e4f125e943d8631a342"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:DUOLM5GFGMSDVWOBAPQLHNWK3F","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Auto-Scaling Heterogeneous Neural Processing Units for Energy and Cost-Efficient LLM Serving","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Jian Huang, Jichuan Chang, Yuqi Xue","submitted_at":"2026-07-17T20:27:26Z","abstract_excerpt":"To meet the ever-increasing computing demands of large language model (LLM) services, modern cloud platforms have widely deployed neural processing units (NPUs). These NPU chips have been developed and evolved at an incredibly fast pace, this inevitably produces heterogeneous compute pools backed by different versions of NPU chips. Unfortunately, due to the lack of system and architecture support for managing NPU heterogeneity in the cloud, it is unclear how to best utilize heterogeneous NPUs to maximize the energy and cost efficiency for LLM services.\n  In this paper, we first conduct a chara"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.16488","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/2607.16488/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":"2026-07-21T00:20:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Gk/X/6e9CjIAqVbuAgb7oTv+FbQ+AzkEipA/XC2P6V7/ngyON0LR2sRyIMDQSzGoaO0b/4/1N4i3vsgBTCYJBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T21:20:58.533575Z"},"content_sha256":"b18a70000030cf7fa03a4ee6c1612dcdeb6c0873aaab63b5112d1ee70e0a9a78","schema_version":"1.0","event_id":"sha256:b18a70000030cf7fa03a4ee6c1612dcdeb6c0873aaab63b5112d1ee70e0a9a78"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DUOLM5GFGMSDVWOBAPQLHNWK3F/bundle.json","state_url":"https://pith.science/pith/DUOLM5GFGMSDVWOBAPQLHNWK3F/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DUOLM5GFGMSDVWOBAPQLHNWK3F/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":"2026-08-03T21:20:58Z","links":{"resolver":"https://pith.science/pith/DUOLM5GFGMSDVWOBAPQLHNWK3F","bundle":"https://pith.science/pith/DUOLM5GFGMSDVWOBAPQLHNWK3F/bundle.json","state":"https://pith.science/pith/DUOLM5GFGMSDVWOBAPQLHNWK3F/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DUOLM5GFGMSDVWOBAPQLHNWK3F/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DUOLM5GFGMSDVWOBAPQLHNWK3F","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":"b4cb20d7130d8fdd2167855c54712651c5986be341f7d0c5485fded1c7e630f9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AR","submitted_at":"2026-07-17T20:27:26Z","title_canon_sha256":"2274af5f55cdeaf5fdc1880ffaf11f59d36cb238da47159c1ac25f98a53125b4"},"schema_version":"1.0","source":{"id":"2607.16488","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.16488","created_at":"2026-07-21T00:20:21Z"},{"alias_kind":"arxiv_version","alias_value":"2607.16488v1","created_at":"2026-07-21T00:20:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.16488","created_at":"2026-07-21T00:20:21Z"},{"alias_kind":"pith_short_12","alias_value":"DUOLM5GFGMSD","created_at":"2026-07-21T00:20:21Z"},{"alias_kind":"pith_short_16","alias_value":"DUOLM5GFGMSDVWOB","created_at":"2026-07-21T00:20:21Z"},{"alias_kind":"pith_short_8","alias_value":"DUOLM5GF","created_at":"2026-07-21T00:20:21Z"}],"graph_snapshots":[{"event_id":"sha256:b18a70000030cf7fa03a4ee6c1612dcdeb6c0873aaab63b5112d1ee70e0a9a78","target":"graph","created_at":"2026-07-21T00:20:21Z","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/2607.16488/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"To meet the ever-increasing computing demands of large language model (LLM) services, modern cloud platforms have widely deployed neural processing units (NPUs). These NPU chips have been developed and evolved at an incredibly fast pace, this inevitably produces heterogeneous compute pools backed by different versions of NPU chips. Unfortunately, due to the lack of system and architecture support for managing NPU heterogeneity in the cloud, it is unclear how to best utilize heterogeneous NPUs to maximize the energy and cost efficiency for LLM services.\n  In this paper, we first conduct a chara","authors_text":"Jian Huang, Jichuan Chang, Yuqi Xue","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AR","submitted_at":"2026-07-17T20:27:26Z","title":"Auto-Scaling Heterogeneous Neural Processing Units for Energy and Cost-Efficient LLM Serving"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.16488","kind":"arxiv","version":1},"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:f758722ef21b0f388510d002a477ce701d39aaf851f01e4f125e943d8631a342","target":"record","created_at":"2026-07-21T00:20:21Z","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":"b4cb20d7130d8fdd2167855c54712651c5986be341f7d0c5485fded1c7e630f9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AR","submitted_at":"2026-07-17T20:27:26Z","title_canon_sha256":"2274af5f55cdeaf5fdc1880ffaf11f59d36cb238da47159c1ac25f98a53125b4"},"schema_version":"1.0","source":{"id":"2607.16488","kind":"arxiv","version":1}},"canonical_sha256":"1d1cb674c533243ad9c103e0b3b6cad9515c2cf7a26cfb25a0e8b417d936e248","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1d1cb674c533243ad9c103e0b3b6cad9515c2cf7a26cfb25a0e8b417d936e248","first_computed_at":"2026-07-21T00:20:21.216478Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-21T00:20:21.216478Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HDDMQaTZZVkbknXiFjZYgCCnEjZersbN05GsKM9mh7rxrMfAirkbje/JwUE80hEJevOb20JHBvYX53iG/+C5Cg==","signature_status":"signed_v1","signed_at":"2026-07-21T00:20:21.217392Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.16488","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f758722ef21b0f388510d002a477ce701d39aaf851f01e4f125e943d8631a342","sha256:b18a70000030cf7fa03a4ee6c1612dcdeb6c0873aaab63b5112d1ee70e0a9a78"],"state_sha256":"83b3c8b7a5a8c52b4f8402e2453c22678736c88522d6cb39f7869980df08499c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xFoKUAV9y11WWOWr50BQClykIVPyQL/0b4ts9vVZTpkI2xy4eRojmlRFD355efWcBmMn7u4P1PxNsPMi1xxSBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T21:20:58.540677Z","bundle_sha256":"e981ae3e491f3ac737e0ddf96324865f7c0450df35d7748b88c9e37d965e3d8e"}}