{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:PO3H6CMWB2ETGL6HR6QCTSM6GC","short_pith_number":"pith:PO3H6CMW","canonical_record":{"source":{"id":"2505.04021","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-05-06T23:38:33Z","cross_cats_sorted":["cs.AI","cs.LG","cs.PF"],"title_canon_sha256":"996b916933bd653a80dfb4f02ee10c8b814ebdfcd4818d634ae6b4daef152027","abstract_canon_sha256":"344dd146af12ff712f8acddfb606640a2411f3f2eb18611d0c02bc4dba473eff"},"schema_version":"1.0"},"canonical_sha256":"7bb67f09960e89332fc78fa029c99e308361d1930735f6f3f88d4834ff5cfa6d","source":{"kind":"arxiv","id":"2505.04021","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.04021","created_at":"2026-06-12T01:08:13Z"},{"alias_kind":"arxiv_version","alias_value":"2505.04021v3","created_at":"2026-06-12T01:08:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.04021","created_at":"2026-06-12T01:08:13Z"},{"alias_kind":"pith_short_12","alias_value":"PO3H6CMWB2ET","created_at":"2026-06-12T01:08:13Z"},{"alias_kind":"pith_short_16","alias_value":"PO3H6CMWB2ETGL6H","created_at":"2026-06-12T01:08:13Z"},{"alias_kind":"pith_short_8","alias_value":"PO3H6CMW","created_at":"2026-06-12T01:08:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:PO3H6CMWB2ETGL6HR6QCTSM6GC","target":"record","payload":{"canonical_record":{"source":{"id":"2505.04021","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-05-06T23:38:33Z","cross_cats_sorted":["cs.AI","cs.LG","cs.PF"],"title_canon_sha256":"996b916933bd653a80dfb4f02ee10c8b814ebdfcd4818d634ae6b4daef152027","abstract_canon_sha256":"344dd146af12ff712f8acddfb606640a2411f3f2eb18611d0c02bc4dba473eff"},"schema_version":"1.0"},"canonical_sha256":"7bb67f09960e89332fc78fa029c99e308361d1930735f6f3f88d4834ff5cfa6d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-12T01:08:13.877776Z","signature_b64":"w0i7pAdoNBTecBIeeIKXAgoBEc3c6/WFKsDtOCr+LROcGerD2XxtCQMwbUiC+Hh6+t+PCs71JxSLYGtefPvCBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7bb67f09960e89332fc78fa029c99e308361d1930735f6f3f88d4834ff5cfa6d","last_reissued_at":"2026-06-12T01:08:13.876716Z","signature_status":"signed_v1","first_computed_at":"2026-06-12T01:08:13.876716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.04021","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-12T01:08:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vO0fh/0qqooDta3frJ4Yh5aRfVROLFE/rQdME5fp567gqdk8gg8AICCZRSB0x/0XXfnvvvK3oC5VQD147TOrBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T13:38:57.746992Z"},"content_sha256":"e403c4856069d706df38fd96dffd529d41572d21a551d9de6881967018063f6a","schema_version":"1.0","event_id":"sha256:e403c4856069d706df38fd96dffd529d41572d21a551d9de6881967018063f6a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:PO3H6CMWB2ETGL6HR6QCTSM6GC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Prism: Cost-Efficient Multi-LLM Serving via GPU Memory Ballooning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.PF"],"primary_cat":"cs.DC","authors_text":"Deepak Vij, Gao Gao, Harry Xu, Jiarong Xing, Junyi Shu, Ke Bao, Mingyuan Ma, Qingda Lu, Shan Yu, Shiyi Cao, Shuo Yang, Xiaoning Ding, Xinyuan Tong, Yangmin Li, Yang Wang, Yichen Wang, Yifan Qiao, Ying Sheng, Yuwei An, Zhiqiang Xie, Zhong Wang","submitted_at":"2025-05-06T23:38:33Z","abstract_excerpt":"Inference providers must maintain availability for many LLMs, including low-volume but essential models, making resource efficiency increasingly important as token prices fall. Analysis of production traces reveals a dynamic bursty-group pattern in which sets of models become active together and shift over time; existing space- and time-sharing approaches lack principled mechanisms to adapt to this variability, forcing trade-offs between SLO adherence and efficiency. We observe that elastic memory allocation can unify spatial and temporal sharing. Based on this insight, we have developed Prism"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.04021","kind":"arxiv","version":3},"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/2505.04021/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":"2026-06-12T01:08:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aMNjadN3EwY/OUC0jFvh5SHe+SuTZP69Nqf13F3nEOmK9eZR6pFWQ3MjrA2oP969F/SI/K3wm2jXQH53FWnGCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T13:38:57.747389Z"},"content_sha256":"6b4327f1eba0427bced9625cf659c9cce4c42e109a492f8a476fc0b8e7160210","schema_version":"1.0","event_id":"sha256:6b4327f1eba0427bced9625cf659c9cce4c42e109a492f8a476fc0b8e7160210"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PO3H6CMWB2ETGL6HR6QCTSM6GC/bundle.json","state_url":"https://pith.science/pith/PO3H6CMWB2ETGL6HR6QCTSM6GC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PO3H6CMWB2ETGL6HR6QCTSM6GC/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":"2026-06-30T13:38:57Z","links":{"resolver":"https://pith.science/pith/PO3H6CMWB2ETGL6HR6QCTSM6GC","bundle":"https://pith.science/pith/PO3H6CMWB2ETGL6HR6QCTSM6GC/bundle.json","state":"https://pith.science/pith/PO3H6CMWB2ETGL6HR6QCTSM6GC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PO3H6CMWB2ETGL6HR6QCTSM6GC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PO3H6CMWB2ETGL6HR6QCTSM6GC","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":"344dd146af12ff712f8acddfb606640a2411f3f2eb18611d0c02bc4dba473eff","cross_cats_sorted":["cs.AI","cs.LG","cs.PF"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-05-06T23:38:33Z","title_canon_sha256":"996b916933bd653a80dfb4f02ee10c8b814ebdfcd4818d634ae6b4daef152027"},"schema_version":"1.0","source":{"id":"2505.04021","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.04021","created_at":"2026-06-12T01:08:13Z"},{"alias_kind":"arxiv_version","alias_value":"2505.04021v3","created_at":"2026-06-12T01:08:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.04021","created_at":"2026-06-12T01:08:13Z"},{"alias_kind":"pith_short_12","alias_value":"PO3H6CMWB2ET","created_at":"2026-06-12T01:08:13Z"},{"alias_kind":"pith_short_16","alias_value":"PO3H6CMWB2ETGL6H","created_at":"2026-06-12T01:08:13Z"},{"alias_kind":"pith_short_8","alias_value":"PO3H6CMW","created_at":"2026-06-12T01:08:13Z"}],"graph_snapshots":[{"event_id":"sha256:6b4327f1eba0427bced9625cf659c9cce4c42e109a492f8a476fc0b8e7160210","target":"graph","created_at":"2026-06-12T01:08:13Z","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/2505.04021/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Inference providers must maintain availability for many LLMs, including low-volume but essential models, making resource efficiency increasingly important as token prices fall. Analysis of production traces reveals a dynamic bursty-group pattern in which sets of models become active together and shift over time; existing space- and time-sharing approaches lack principled mechanisms to adapt to this variability, forcing trade-offs between SLO adherence and efficiency. We observe that elastic memory allocation can unify spatial and temporal sharing. Based on this insight, we have developed Prism","authors_text":"Deepak Vij, Gao Gao, Harry Xu, Jiarong Xing, Junyi Shu, Ke Bao, Mingyuan Ma, Qingda Lu, Shan Yu, Shiyi Cao, Shuo Yang, Xiaoning Ding, Xinyuan Tong, Yangmin Li, Yang Wang, Yichen Wang, Yifan Qiao, Ying Sheng, Yuwei An, Zhiqiang Xie, Zhong Wang","cross_cats":["cs.AI","cs.LG","cs.PF"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-05-06T23:38:33Z","title":"Prism: Cost-Efficient Multi-LLM Serving via GPU Memory Ballooning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.04021","kind":"arxiv","version":3},"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:e403c4856069d706df38fd96dffd529d41572d21a551d9de6881967018063f6a","target":"record","created_at":"2026-06-12T01:08:13Z","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":"344dd146af12ff712f8acddfb606640a2411f3f2eb18611d0c02bc4dba473eff","cross_cats_sorted":["cs.AI","cs.LG","cs.PF"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-05-06T23:38:33Z","title_canon_sha256":"996b916933bd653a80dfb4f02ee10c8b814ebdfcd4818d634ae6b4daef152027"},"schema_version":"1.0","source":{"id":"2505.04021","kind":"arxiv","version":3}},"canonical_sha256":"7bb67f09960e89332fc78fa029c99e308361d1930735f6f3f88d4834ff5cfa6d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7bb67f09960e89332fc78fa029c99e308361d1930735f6f3f88d4834ff5cfa6d","first_computed_at":"2026-06-12T01:08:13.876716Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-12T01:08:13.876716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"w0i7pAdoNBTecBIeeIKXAgoBEc3c6/WFKsDtOCr+LROcGerD2XxtCQMwbUiC+Hh6+t+PCs71JxSLYGtefPvCBg==","signature_status":"signed_v1","signed_at":"2026-06-12T01:08:13.877776Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.04021","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e403c4856069d706df38fd96dffd529d41572d21a551d9de6881967018063f6a","sha256:6b4327f1eba0427bced9625cf659c9cce4c42e109a492f8a476fc0b8e7160210"],"state_sha256":"dfbcac2b84866dd93490b2549dc04829f8cfaa58bdb92c6732ed93d3662fccbe"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FrM+g6CJVZjpmRaaGWOH0N4UR82FDm+5lQmjAzEZqVoJT5zzwb+kMJoMEZfjrUyyaleo++RxWREsg7TnwUZUBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-30T13:38:57.749438Z","bundle_sha256":"3f64c7c3e517d058bd49ebc8be56939cfc4b01b12a547943b72c73f67e33b9ee"}}