{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FVFHMEXPVBXLP4NYFCHB2UMEPF","short_pith_number":"pith:FVFHMEXP","schema_version":"1.0","canonical_sha256":"2d4a7612efa86eb7f1b8288e1d5184796d736037ad375f592477c66135407ba6","source":{"kind":"arxiv","id":"2607.11136","version":1},"attestation_state":"computed","paper":{"title":"Xema: Efficient Diffusion Serving through Fine-Grained Memory Management and Auto-Configuration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Guangyu Xiang, Lin Zhang, Shaohuai Shi, Suyi Li, Xiaowen Chu, Xueze Kang, Yuxin Wang","submitted_at":"2026-07-13T06:12:41Z","abstract_excerpt":"Diffusion models are increasingly deployed as production visual-generation services, where serving high-resolution image and long video generation is often limited by GPU memory. Popular memory-saving techniques such as weight offloading, sharding, and VAE slicing are often not practical because they tend to introduce significant performance overhead. In this paper, we present Xema, a memory-efficient diffusion serving system that exploits predictable tensor lifetimes for trace-guided memory optimization. For each request template, Xema derives an offline memory trace to identify short memory-"},"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":"2607.11136","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2026-07-13T06:12:41Z","cross_cats_sorted":[],"title_canon_sha256":"4065997d145e5cd03360ba430bf4d20df8ef22ea6027b11b8e4ce1ec69553799","abstract_canon_sha256":"40d5da49fb38fb3d6263308fb0ef53f4edd40954e5eda7126b8ef2773cb5bef4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T01:22:18.307052Z","signature_b64":"oywuxUe/cJxF9WseEoBMob9p9OIkOC9IfsvY88bJJqB+TbhbE69WYcHXmtsW3Mgl41Rgau8d08eR1I6TD93BDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2d4a7612efa86eb7f1b8288e1d5184796d736037ad375f592477c66135407ba6","last_reissued_at":"2026-07-14T01:22:18.306195Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T01:22:18.306195Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Xema: Efficient Diffusion Serving through Fine-Grained Memory Management and Auto-Configuration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Guangyu Xiang, Lin Zhang, Shaohuai Shi, Suyi Li, Xiaowen Chu, Xueze Kang, Yuxin Wang","submitted_at":"2026-07-13T06:12:41Z","abstract_excerpt":"Diffusion models are increasingly deployed as production visual-generation services, where serving high-resolution image and long video generation is often limited by GPU memory. Popular memory-saving techniques such as weight offloading, sharding, and VAE slicing are often not practical because they tend to introduce significant performance overhead. In this paper, we present Xema, a memory-efficient diffusion serving system that exploits predictable tensor lifetimes for trace-guided memory optimization. For each request template, Xema derives an offline memory trace to identify short memory-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.11136","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/2607.11136/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":"2607.11136","created_at":"2026-07-14T01:22:18.306649+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.11136v1","created_at":"2026-07-14T01:22:18.306649+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.11136","created_at":"2026-07-14T01:22:18.306649+00:00"},{"alias_kind":"pith_short_12","alias_value":"FVFHMEXPVBXL","created_at":"2026-07-14T01:22:18.306649+00:00"},{"alias_kind":"pith_short_16","alias_value":"FVFHMEXPVBXLP4NY","created_at":"2026-07-14T01:22:18.306649+00:00"},{"alias_kind":"pith_short_8","alias_value":"FVFHMEXP","created_at":"2026-07-14T01:22:18.306649+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/FVFHMEXPVBXLP4NYFCHB2UMEPF","json":"https://pith.science/pith/FVFHMEXPVBXLP4NYFCHB2UMEPF.json","graph_json":"https://pith.science/api/pith-number/FVFHMEXPVBXLP4NYFCHB2UMEPF/graph.json","events_json":"https://pith.science/api/pith-number/FVFHMEXPVBXLP4NYFCHB2UMEPF/events.json","paper":"https://pith.science/paper/FVFHMEXP"},"agent_actions":{"view_html":"https://pith.science/pith/FVFHMEXPVBXLP4NYFCHB2UMEPF","download_json":"https://pith.science/pith/FVFHMEXPVBXLP4NYFCHB2UMEPF.json","view_paper":"https://pith.science/paper/FVFHMEXP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.11136&json=true","fetch_graph":"https://pith.science/api/pith-number/FVFHMEXPVBXLP4NYFCHB2UMEPF/graph.json","fetch_events":"https://pith.science/api/pith-number/FVFHMEXPVBXLP4NYFCHB2UMEPF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FVFHMEXPVBXLP4NYFCHB2UMEPF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FVFHMEXPVBXLP4NYFCHB2UMEPF/action/storage_attestation","attest_author":"https://pith.science/pith/FVFHMEXPVBXLP4NYFCHB2UMEPF/action/author_attestation","sign_citation":"https://pith.science/pith/FVFHMEXPVBXLP4NYFCHB2UMEPF/action/citation_signature","submit_replication":"https://pith.science/pith/FVFHMEXPVBXLP4NYFCHB2UMEPF/action/replication_record"}},"created_at":"2026-07-14T01:22:18.306649+00:00","updated_at":"2026-07-14T01:22:18.306649+00:00"}