{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ZMBCRUQ5P72MZCARHKPRFRXRCU","short_pith_number":"pith:ZMBCRUQ5","canonical_record":{"source":{"id":"2509.01229","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-09-01T08:16:20Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"cb3b26fe8d4b313c989cf4faf89b1cb10c93d64370e3bc0deaf95348059d6eb1","abstract_canon_sha256":"d2edbcb9a5299f9c563f6ff16e7aab6643d85e4dd0a1b412a61a344058d1e30b"},"schema_version":"1.0"},"canonical_sha256":"cb0228d21d7ff4cc88113a9f12c6f1152b384050483679bb94c06f3619e6b31e","source":{"kind":"arxiv","id":"2509.01229","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.01229","created_at":"2026-07-05T12:02:56Z"},{"alias_kind":"arxiv_version","alias_value":"2509.01229v1","created_at":"2026-07-05T12:02:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.01229","created_at":"2026-07-05T12:02:56Z"},{"alias_kind":"pith_short_12","alias_value":"ZMBCRUQ5P72M","created_at":"2026-07-05T12:02:56Z"},{"alias_kind":"pith_short_16","alias_value":"ZMBCRUQ5P72MZCAR","created_at":"2026-07-05T12:02:56Z"},{"alias_kind":"pith_short_8","alias_value":"ZMBCRUQ5","created_at":"2026-07-05T12:02:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ZMBCRUQ5P72MZCARHKPRFRXRCU","target":"record","payload":{"canonical_record":{"source":{"id":"2509.01229","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-09-01T08:16:20Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"cb3b26fe8d4b313c989cf4faf89b1cb10c93d64370e3bc0deaf95348059d6eb1","abstract_canon_sha256":"d2edbcb9a5299f9c563f6ff16e7aab6643d85e4dd0a1b412a61a344058d1e30b"},"schema_version":"1.0"},"canonical_sha256":"cb0228d21d7ff4cc88113a9f12c6f1152b384050483679bb94c06f3619e6b31e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:02:56.490919Z","signature_b64":"vOBzBXC7gZiO5rQ9S6skgdpiQ6bCS997ditE6q4ChVhyAXsLA5ZAJVHroAC+tXkX1moeFCk6i/Du1jmeowvICg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cb0228d21d7ff4cc88113a9f12c6f1152b384050483679bb94c06f3619e6b31e","last_reissued_at":"2026-07-05T12:02:56.490433Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:02:56.490433Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.01229","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":"2026-07-05T12:02:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bpl+XwPALMtPpaM7kxO1PVZ4EaWLsYKvGtuGnB/Os+Hx7nAhXLMy6crQDJPx4HJiecZGvUjjH5sRj2EFT7S9BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T10:51:49.151997Z"},"content_sha256":"cda258f5b2c633a6b4c7df48477b3fe72b5cb0608b0e8602ecf072b684ce74f6","schema_version":"1.0","event_id":"sha256:cda258f5b2c633a6b4c7df48477b3fe72b5cb0608b0e8602ecf072b684ce74f6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ZMBCRUQ5P72MZCARHKPRFRXRCU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LiquidGEMM: Hardware-Efficient W4A8 GEMM Kernel for High-Performance LLM Serving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.DC","authors_text":"Bowen Xiao, Chengquan Jiang, Huanqi Hu, Jianian Yin, Minyi Guo, Shixuan Sun, Weiqi Xu, Xiang Luo, Xiaoying Jia, Xin Liu, Zhexi Zhang","submitted_at":"2025-09-01T08:16:20Z","abstract_excerpt":"Quantization is a critical technique for accelerating LLM inference by reducing memory footprint and improving computational efficiency. Among various schemes, 4-bit weight and 8-bit activation quantization (W4A8) offers a strong balance between accuracy and performance. However, existing W4A8 GEMM kernels fall short in practice due to inefficient dequantization on CUDA Cores, which cannot keep pace with the high throughput of Tensor Cores. In this paper, we present LiquidGEMM, a hardware-efficient W4A8 GEMM kernel for efficient LLM serving. LiquidGEMM designs two key techniques: LiquidQuant, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.01229","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/2509.01229/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":"2026-07-05T12:02:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KBPy6Kswtmxk8vR8/KU/UOanssIceMEBxqSWRPmFr68Hs9HKV8ECscfSyIRPXfWIdCcdJfbEHIraAJLOYB14CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T10:51:49.152486Z"},"content_sha256":"35792a4cd94143e3b3e918bf31a7b57638334c0bdb2d8cd1e02eafc69ab73d80","schema_version":"1.0","event_id":"sha256:35792a4cd94143e3b3e918bf31a7b57638334c0bdb2d8cd1e02eafc69ab73d80"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZMBCRUQ5P72MZCARHKPRFRXRCU/bundle.json","state_url":"https://pith.science/pith/ZMBCRUQ5P72MZCARHKPRFRXRCU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZMBCRUQ5P72MZCARHKPRFRXRCU/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":"2026-08-06T10:51:49Z","links":{"resolver":"https://pith.science/pith/ZMBCRUQ5P72MZCARHKPRFRXRCU","bundle":"https://pith.science/pith/ZMBCRUQ5P72MZCARHKPRFRXRCU/bundle.json","state":"https://pith.science/pith/ZMBCRUQ5P72MZCARHKPRFRXRCU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZMBCRUQ5P72MZCARHKPRFRXRCU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZMBCRUQ5P72MZCARHKPRFRXRCU","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":"d2edbcb9a5299f9c563f6ff16e7aab6643d85e4dd0a1b412a61a344058d1e30b","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-09-01T08:16:20Z","title_canon_sha256":"cb3b26fe8d4b313c989cf4faf89b1cb10c93d64370e3bc0deaf95348059d6eb1"},"schema_version":"1.0","source":{"id":"2509.01229","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.01229","created_at":"2026-07-05T12:02:56Z"},{"alias_kind":"arxiv_version","alias_value":"2509.01229v1","created_at":"2026-07-05T12:02:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.01229","created_at":"2026-07-05T12:02:56Z"},{"alias_kind":"pith_short_12","alias_value":"ZMBCRUQ5P72M","created_at":"2026-07-05T12:02:56Z"},{"alias_kind":"pith_short_16","alias_value":"ZMBCRUQ5P72MZCAR","created_at":"2026-07-05T12:02:56Z"},{"alias_kind":"pith_short_8","alias_value":"ZMBCRUQ5","created_at":"2026-07-05T12:02:56Z"}],"graph_snapshots":[{"event_id":"sha256:35792a4cd94143e3b3e918bf31a7b57638334c0bdb2d8cd1e02eafc69ab73d80","target":"graph","created_at":"2026-07-05T12:02:56Z","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/2509.01229/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Quantization is a critical technique for accelerating LLM inference by reducing memory footprint and improving computational efficiency. Among various schemes, 4-bit weight and 8-bit activation quantization (W4A8) offers a strong balance between accuracy and performance. However, existing W4A8 GEMM kernels fall short in practice due to inefficient dequantization on CUDA Cores, which cannot keep pace with the high throughput of Tensor Cores. In this paper, we present LiquidGEMM, a hardware-efficient W4A8 GEMM kernel for efficient LLM serving. LiquidGEMM designs two key techniques: LiquidQuant, ","authors_text":"Bowen Xiao, Chengquan Jiang, Huanqi Hu, Jianian Yin, Minyi Guo, Shixuan Sun, Weiqi Xu, Xiang Luo, Xiaoying Jia, Xin Liu, Zhexi Zhang","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-09-01T08:16:20Z","title":"LiquidGEMM: Hardware-Efficient W4A8 GEMM Kernel for High-Performance LLM Serving"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.01229","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:cda258f5b2c633a6b4c7df48477b3fe72b5cb0608b0e8602ecf072b684ce74f6","target":"record","created_at":"2026-07-05T12:02:56Z","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":"d2edbcb9a5299f9c563f6ff16e7aab6643d85e4dd0a1b412a61a344058d1e30b","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-09-01T08:16:20Z","title_canon_sha256":"cb3b26fe8d4b313c989cf4faf89b1cb10c93d64370e3bc0deaf95348059d6eb1"},"schema_version":"1.0","source":{"id":"2509.01229","kind":"arxiv","version":1}},"canonical_sha256":"cb0228d21d7ff4cc88113a9f12c6f1152b384050483679bb94c06f3619e6b31e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cb0228d21d7ff4cc88113a9f12c6f1152b384050483679bb94c06f3619e6b31e","first_computed_at":"2026-07-05T12:02:56.490433Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:02:56.490433Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vOBzBXC7gZiO5rQ9S6skgdpiQ6bCS997ditE6q4ChVhyAXsLA5ZAJVHroAC+tXkX1moeFCk6i/Du1jmeowvICg==","signature_status":"signed_v1","signed_at":"2026-07-05T12:02:56.490919Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.01229","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cda258f5b2c633a6b4c7df48477b3fe72b5cb0608b0e8602ecf072b684ce74f6","sha256:35792a4cd94143e3b3e918bf31a7b57638334c0bdb2d8cd1e02eafc69ab73d80"],"state_sha256":"61e67e2a17828c9a9a3b94be83499d58048c0a93af1f83abb58677021b70bd15"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"laeXwW7CgSpy2S1GTjcz+op6iatlmYbat5lz2uvri8jgHJlselSV4WXfL49FCznANiAMcOPtZzaJFZJsctRxAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T10:51:49.155943Z","bundle_sha256":"251f42a22e1733d262ae216aeaf02e119dddf97c1f5ad76d012b996d4b7367e8"}}