{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PFLLN7U24U4W65HSRC2L3ZSA66","short_pith_number":"pith:PFLLN7U2","schema_version":"1.0","canonical_sha256":"7956b6fe9ae5396f74f288b4bde640f7821464824b189e153568b8e9f622a509","source":{"kind":"arxiv","id":"2508.10395","version":1},"attestation_state":"computed","paper":{"title":"XQuant: Breaking the Memory Wall for LLM Inference with KV Cache Rematerialization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aditya Tomar, Amir Gholami, Coleman Hooper, Haocheng Xi, Kurt Keutzer, Luca Manolache, Michael W. Mahoney, Minjae Lee, Rishabh Tiwari, Wonjun Kang","submitted_at":"2025-08-14T06:52:38Z","abstract_excerpt":"Although LLM inference has emerged as a critical workload for many downstream applications, efficiently inferring LLMs is challenging due to the substantial memory footprint and bandwidth requirements. In parallel, compute capabilities have steadily outpaced both memory capacity and bandwidth over the last few decades, a trend that remains evident in modern GPU hardware and exacerbates the challenge of LLM inference. As such, new algorithms are emerging that trade increased computation for reduced memory operations. To that end, we present XQuant, which takes advantage of this trend, enabling "},"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":"2508.10395","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-08-14T06:52:38Z","cross_cats_sorted":[],"title_canon_sha256":"c8ce0c84c02bbd0acebcf4e3b3706a192b569597014998947d9e9b644dee6154","abstract_canon_sha256":"f576b028c36dafcbbb229532befb4c74a7cccbb2a3fa883da6f47700b0c86e35"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:55.285436Z","signature_b64":"zAtAhP4TgRr+ActQjDr8xIdZf2I/0b/DQIJy0wR4f78dj8GYDgJEJygvOQoE+mVuU9tGyxQBUM+VDYKIiuytAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7956b6fe9ae5396f74f288b4bde640f7821464824b189e153568b8e9f622a509","last_reissued_at":"2026-07-05T11:53:55.284894Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:55.284894Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"XQuant: Breaking the Memory Wall for LLM Inference with KV Cache Rematerialization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aditya Tomar, Amir Gholami, Coleman Hooper, Haocheng Xi, Kurt Keutzer, Luca Manolache, Michael W. Mahoney, Minjae Lee, Rishabh Tiwari, Wonjun Kang","submitted_at":"2025-08-14T06:52:38Z","abstract_excerpt":"Although LLM inference has emerged as a critical workload for many downstream applications, efficiently inferring LLMs is challenging due to the substantial memory footprint and bandwidth requirements. In parallel, compute capabilities have steadily outpaced both memory capacity and bandwidth over the last few decades, a trend that remains evident in modern GPU hardware and exacerbates the challenge of LLM inference. As such, new algorithms are emerging that trade increased computation for reduced memory operations. To that end, we present XQuant, which takes advantage of this trend, enabling "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.10395","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/2508.10395/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":"2508.10395","created_at":"2026-07-05T11:53:55.284964+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.10395v1","created_at":"2026-07-05T11:53:55.284964+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.10395","created_at":"2026-07-05T11:53:55.284964+00:00"},{"alias_kind":"pith_short_12","alias_value":"PFLLN7U24U4W","created_at":"2026-07-05T11:53:55.284964+00:00"},{"alias_kind":"pith_short_16","alias_value":"PFLLN7U24U4W65HS","created_at":"2026-07-05T11:53:55.284964+00:00"},{"alias_kind":"pith_short_8","alias_value":"PFLLN7U2","created_at":"2026-07-05T11:53:55.284964+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/PFLLN7U24U4W65HSRC2L3ZSA66","json":"https://pith.science/pith/PFLLN7U24U4W65HSRC2L3ZSA66.json","graph_json":"https://pith.science/api/pith-number/PFLLN7U24U4W65HSRC2L3ZSA66/graph.json","events_json":"https://pith.science/api/pith-number/PFLLN7U24U4W65HSRC2L3ZSA66/events.json","paper":"https://pith.science/paper/PFLLN7U2"},"agent_actions":{"view_html":"https://pith.science/pith/PFLLN7U24U4W65HSRC2L3ZSA66","download_json":"https://pith.science/pith/PFLLN7U24U4W65HSRC2L3ZSA66.json","view_paper":"https://pith.science/paper/PFLLN7U2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.10395&json=true","fetch_graph":"https://pith.science/api/pith-number/PFLLN7U24U4W65HSRC2L3ZSA66/graph.json","fetch_events":"https://pith.science/api/pith-number/PFLLN7U24U4W65HSRC2L3ZSA66/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PFLLN7U24U4W65HSRC2L3ZSA66/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PFLLN7U24U4W65HSRC2L3ZSA66/action/storage_attestation","attest_author":"https://pith.science/pith/PFLLN7U24U4W65HSRC2L3ZSA66/action/author_attestation","sign_citation":"https://pith.science/pith/PFLLN7U24U4W65HSRC2L3ZSA66/action/citation_signature","submit_replication":"https://pith.science/pith/PFLLN7U24U4W65HSRC2L3ZSA66/action/replication_record"}},"created_at":"2026-07-05T11:53:55.284964+00:00","updated_at":"2026-07-05T11:53:55.284964+00:00"}