{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:OTAJLRKWGWPABL6SIZK3MQKK3V","short_pith_number":"pith:OTAJLRKW","canonical_record":{"source":{"id":"2505.19342","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-25T22:16:59Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c854ac4c837e8b001a185a257248b9dd7fcc3a4117aed362d098c52304de2996","abstract_canon_sha256":"665ffff21f65346ad1fa276b5bf59d88febc7a2fe278dcfd60d08e7dc6c68650"},"schema_version":"1.0"},"canonical_sha256":"74c095c556359e00afd24655b6414add74a8e59fa8691ee43cd2ebc82e9ffa52","source":{"kind":"arxiv","id":"2505.19342","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19342","created_at":"2026-05-28T01:04:28Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19342v2","created_at":"2026-05-28T01:04:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19342","created_at":"2026-05-28T01:04:28Z"},{"alias_kind":"pith_short_12","alias_value":"OTAJLRKWGWPA","created_at":"2026-05-28T01:04:28Z"},{"alias_kind":"pith_short_16","alias_value":"OTAJLRKWGWPABL6S","created_at":"2026-05-28T01:04:28Z"},{"alias_kind":"pith_short_8","alias_value":"OTAJLRKW","created_at":"2026-05-28T01:04:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:OTAJLRKWGWPABL6SIZK3MQKK3V","target":"record","payload":{"canonical_record":{"source":{"id":"2505.19342","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-25T22:16:59Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c854ac4c837e8b001a185a257248b9dd7fcc3a4117aed362d098c52304de2996","abstract_canon_sha256":"665ffff21f65346ad1fa276b5bf59d88febc7a2fe278dcfd60d08e7dc6c68650"},"schema_version":"1.0"},"canonical_sha256":"74c095c556359e00afd24655b6414add74a8e59fa8691ee43cd2ebc82e9ffa52","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:28.301965Z","signature_b64":"Zb8+bcheBF3J29Z10UDZUSiOMfxMJ5/Gu3eAQoArE0K/v4kgfezRVNOzRLd68nEdxaLSoKqGkLEuOLCQB2N6BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"74c095c556359e00afd24655b6414add74a8e59fa8691ee43cd2ebc82e9ffa52","last_reissued_at":"2026-05-28T01:04:28.301381Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:28.301381Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.19342","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-28T01:04:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FLi5zkgc9r3YSWo8uTpYwcKe1/42LfgPW+5eifVEWqQNdLWzuntKvv82VUbo82bNdM1ZT/hQLA91Zz+gux/UAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T01:38:16.720112Z"},"content_sha256":"afd1a49e1ae892d62de69afe58a708c3e3035ea9c491de038276b607db6e5bb2","schema_version":"1.0","event_id":"sha256:afd1a49e1ae892d62de69afe58a708c3e3035ea9c491de038276b607db6e5bb2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:OTAJLRKWGWPABL6SIZK3MQKK3V","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ASTRA: Communication-Efficient Acceleration for Multi-Device Transformer Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Deepak Ganesan, Hui Guan, Lijun Zhang, Xiao Liu","submitted_at":"2025-05-25T22:16:59Z","abstract_excerpt":"Multi-device inference can reduce Transformer latency by parallelizing computation. However, existing methods require high inter-device bandwidth, making them impractical for bandwidth-constrained environments. We present ASTRA, a communication-efficient framework that integrates sequence parallelism with mixed-precision attention, where non-local token embeddings are transmitted as low-bit vector-quantized codes while local attention remains full precision. To preserve accuracy under aggressive compression, ASTRA introduces Noise-Augmented Quantization and Distributed Class Tokens. Across vis"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19342","kind":"arxiv","version":2},"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/2505.19342/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":"2026-05-28T01:04:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"z9q3ika8G87qqtBUgeX4+4pDH+fvN/2BAkHjXm/EZinsmLeSM6s/mU4xB2N2hUo2TNKKZpe/h2LJH6qmFtbBBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T01:38:16.720792Z"},"content_sha256":"7f27068cf2b49023987a58f707325fc5d08b121e96e08e0c4c3ef9d7194ca512","schema_version":"1.0","event_id":"sha256:7f27068cf2b49023987a58f707325fc5d08b121e96e08e0c4c3ef9d7194ca512"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OTAJLRKWGWPABL6SIZK3MQKK3V/bundle.json","state_url":"https://pith.science/pith/OTAJLRKWGWPABL6SIZK3MQKK3V/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OTAJLRKWGWPABL6SIZK3MQKK3V/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":"2026-05-31T01:38:16Z","links":{"resolver":"https://pith.science/pith/OTAJLRKWGWPABL6SIZK3MQKK3V","bundle":"https://pith.science/pith/OTAJLRKWGWPABL6SIZK3MQKK3V/bundle.json","state":"https://pith.science/pith/OTAJLRKWGWPABL6SIZK3MQKK3V/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OTAJLRKWGWPABL6SIZK3MQKK3V/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OTAJLRKWGWPABL6SIZK3MQKK3V","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":"665ffff21f65346ad1fa276b5bf59d88febc7a2fe278dcfd60d08e7dc6c68650","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-25T22:16:59Z","title_canon_sha256":"c854ac4c837e8b001a185a257248b9dd7fcc3a4117aed362d098c52304de2996"},"schema_version":"1.0","source":{"id":"2505.19342","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19342","created_at":"2026-05-28T01:04:28Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19342v2","created_at":"2026-05-28T01:04:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19342","created_at":"2026-05-28T01:04:28Z"},{"alias_kind":"pith_short_12","alias_value":"OTAJLRKWGWPA","created_at":"2026-05-28T01:04:28Z"},{"alias_kind":"pith_short_16","alias_value":"OTAJLRKWGWPABL6S","created_at":"2026-05-28T01:04:28Z"},{"alias_kind":"pith_short_8","alias_value":"OTAJLRKW","created_at":"2026-05-28T01:04:28Z"}],"graph_snapshots":[{"event_id":"sha256:7f27068cf2b49023987a58f707325fc5d08b121e96e08e0c4c3ef9d7194ca512","target":"graph","created_at":"2026-05-28T01:04:28Z","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/2505.19342/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multi-device inference can reduce Transformer latency by parallelizing computation. However, existing methods require high inter-device bandwidth, making them impractical for bandwidth-constrained environments. We present ASTRA, a communication-efficient framework that integrates sequence parallelism with mixed-precision attention, where non-local token embeddings are transmitted as low-bit vector-quantized codes while local attention remains full precision. To preserve accuracy under aggressive compression, ASTRA introduces Noise-Augmented Quantization and Distributed Class Tokens. Across vis","authors_text":"Deepak Ganesan, Hui Guan, Lijun Zhang, Xiao Liu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-25T22:16:59Z","title":"ASTRA: Communication-Efficient Acceleration for Multi-Device Transformer Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19342","kind":"arxiv","version":2},"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:afd1a49e1ae892d62de69afe58a708c3e3035ea9c491de038276b607db6e5bb2","target":"record","created_at":"2026-05-28T01:04:28Z","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":"665ffff21f65346ad1fa276b5bf59d88febc7a2fe278dcfd60d08e7dc6c68650","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-25T22:16:59Z","title_canon_sha256":"c854ac4c837e8b001a185a257248b9dd7fcc3a4117aed362d098c52304de2996"},"schema_version":"1.0","source":{"id":"2505.19342","kind":"arxiv","version":2}},"canonical_sha256":"74c095c556359e00afd24655b6414add74a8e59fa8691ee43cd2ebc82e9ffa52","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"74c095c556359e00afd24655b6414add74a8e59fa8691ee43cd2ebc82e9ffa52","first_computed_at":"2026-05-28T01:04:28.301381Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:04:28.301381Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Zb8+bcheBF3J29Z10UDZUSiOMfxMJ5/Gu3eAQoArE0K/v4kgfezRVNOzRLd68nEdxaLSoKqGkLEuOLCQB2N6BA==","signature_status":"signed_v1","signed_at":"2026-05-28T01:04:28.301965Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.19342","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:afd1a49e1ae892d62de69afe58a708c3e3035ea9c491de038276b607db6e5bb2","sha256:7f27068cf2b49023987a58f707325fc5d08b121e96e08e0c4c3ef9d7194ca512"],"state_sha256":"10897fc844785967ded3a8a945b023420e02cea9c8459f94c58df76e70960de5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"f1fVt5LgUcZbzKygMJIIOQ7wfkBphYqGzdvhsdwAyS6UMNLoZf7P/VxvI2f4smgMcCIy9HMly5idvJMurbuzCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T01:38:16.725642Z","bundle_sha256":"c70ba41f4627fcce5c7a0998baef363d68f28dc5fe352086087b4a5c78af73b5"}}