{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:AKOWOTOE5O3RYQ2ZBF6AR7MRNS","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":"e6d6c5a597b0e3725db10f4c3adfa7d6d0a7a18ef7e290f4e8c00a90e4a9f8ae","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2026-06-07T18:01:55Z","title_canon_sha256":"d357a9894e03ec24bcc298c150760570e00af4c9bbbf601edf7cc1a65f3f9c30"},"schema_version":"1.0","source":{"id":"2606.08761","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.08761","created_at":"2026-06-09T02:07:37Z"},{"alias_kind":"arxiv_version","alias_value":"2606.08761v1","created_at":"2026-06-09T02:07:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.08761","created_at":"2026-06-09T02:07:37Z"},{"alias_kind":"pith_short_12","alias_value":"AKOWOTOE5O3R","created_at":"2026-06-09T02:07:37Z"},{"alias_kind":"pith_short_16","alias_value":"AKOWOTOE5O3RYQ2Z","created_at":"2026-06-09T02:07:37Z"},{"alias_kind":"pith_short_8","alias_value":"AKOWOTOE","created_at":"2026-06-09T02:07:37Z"}],"graph_snapshots":[{"event_id":"sha256:a2c800d14509d1a66fd29e9d1ab27b2d21620c6fdba55103f4edc31079417fb4","target":"graph","created_at":"2026-06-09T02:07:37Z","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/2606.08761/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"W4A4 quantization promises full utilization of INT4 Tensor Cores, yet group dequantization overhead on CUDA Cores has driven existing systems to mixed-precision fallbacks. We present the first systematic study of how intra-SM compute balance governs this bottleneck. Through controlled benchmarks across four GPUs from Ampere and Ada architectures, we identify the Tensor Cores to CUDA Cores throughput ratio ($\\rho$) as the primary hardware indicator: the W4A4-g128 kernel yields $2.0$--$2.5\\times$ speedup on RTX~3090 ($\\rho=16$) yet degrades to $0.43$--$0.47\\times$ on A100 ($\\rho=64$) in compute-","authors_text":"Christoph Meinel, Haojin Yang, Hong Guo, Jona Otholt, Nianhui Guo, Weixing Wang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2026-06-07T18:01:55Z","title":"APEX4: Efficient Pure W4A4 LLM Inference via Intra-SM Compute Rebalancing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.08761","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:ede540a59a719cd627f715787220b1ab44107cf700aaac22aebca23304969257","target":"record","created_at":"2026-06-09T02:07:37Z","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":"e6d6c5a597b0e3725db10f4c3adfa7d6d0a7a18ef7e290f4e8c00a90e4a9f8ae","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2026-06-07T18:01:55Z","title_canon_sha256":"d357a9894e03ec24bcc298c150760570e00af4c9bbbf601edf7cc1a65f3f9c30"},"schema_version":"1.0","source":{"id":"2606.08761","kind":"arxiv","version":1}},"canonical_sha256":"029d674dc4ebb71c4359097c08fd916c87c493a83924c3a74eb9192d0a023a1b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"029d674dc4ebb71c4359097c08fd916c87c493a83924c3a74eb9192d0a023a1b","first_computed_at":"2026-06-09T02:07:37.874710Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-09T02:07:37.874710Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Fgcs/4K9hwz8T7RASyHFxyd3VHBl3p/Wiv/6h3uFfZ+bGT8mOIDAYeOGItAXedjKgGF7ktXGfmRHTAxHXnZvBw==","signature_status":"signed_v1","signed_at":"2026-06-09T02:07:37.875611Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.08761","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ede540a59a719cd627f715787220b1ab44107cf700aaac22aebca23304969257","sha256:a2c800d14509d1a66fd29e9d1ab27b2d21620c6fdba55103f4edc31079417fb4"],"state_sha256":"04690e6c83008125aa00c943306d4fddccf3af00506148a08f23653de49e61c5"}