{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:NBIWZ3MIKYELQEWCVGJ2FAYYI2","short_pith_number":"pith:NBIWZ3MI","canonical_record":{"source":{"id":"2306.01076","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-01T18:32:08Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"912b0613b7a52f7cfe18f6a327193ce5680143db3dc35f7695cd8fd2e9ee4e53","abstract_canon_sha256":"3c4cf910b5c89c6a7c8c70771588e578281764cb1dc4d3c082591d4d4215e2e3"},"schema_version":"1.0"},"canonical_sha256":"68516ced885608b812c2a993a2831846a64d98335f30cc712194f2ee182b028d","source":{"kind":"arxiv","id":"2306.01076","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.01076","created_at":"2026-07-05T09:12:53Z"},{"alias_kind":"arxiv_version","alias_value":"2306.01076v2","created_at":"2026-07-05T09:12:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.01076","created_at":"2026-07-05T09:12:53Z"},{"alias_kind":"pith_short_12","alias_value":"NBIWZ3MIKYEL","created_at":"2026-07-05T09:12:53Z"},{"alias_kind":"pith_short_16","alias_value":"NBIWZ3MIKYELQEWC","created_at":"2026-07-05T09:12:53Z"},{"alias_kind":"pith_short_8","alias_value":"NBIWZ3MI","created_at":"2026-07-05T09:12:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:NBIWZ3MIKYELQEWCVGJ2FAYYI2","target":"record","payload":{"canonical_record":{"source":{"id":"2306.01076","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-01T18:32:08Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"912b0613b7a52f7cfe18f6a327193ce5680143db3dc35f7695cd8fd2e9ee4e53","abstract_canon_sha256":"3c4cf910b5c89c6a7c8c70771588e578281764cb1dc4d3c082591d4d4215e2e3"},"schema_version":"1.0"},"canonical_sha256":"68516ced885608b812c2a993a2831846a64d98335f30cc712194f2ee182b028d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:12:53.390576Z","signature_b64":"hscnDMlLapI60jUfhLEmdEkq/IRPQDumAkuVe7VUQxSIM5PECiSpJuptDq+cViyFdWXgKL9IlQUQK5vYZ4NkDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"68516ced885608b812c2a993a2831846a64d98335f30cc712194f2ee182b028d","last_reissued_at":"2026-07-05T09:12:53.390092Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:12:53.390092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2306.01076","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":"2026-07-05T09:12:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3/JCRGkp1gCAXsu4MBBcjieUngXhMeoOb28Avk7WSoOYpAe187uULo6wMg5i8hrevqCvuXuFsBZAzvcCHH7mAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T12:40:36.967498Z"},"content_sha256":"9ec1f787d98dad1d2595e0eb234a4b6240b5a6b3cfe20c42bc25238f95c63d50","schema_version":"1.0","event_id":"sha256:9ec1f787d98dad1d2595e0eb234a4b6240b5a6b3cfe20c42bc25238f95c63d50"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:NBIWZ3MIKYELQEWCVGJ2FAYYI2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Quantization-Aware and Tensor-Compressed Training of Transformers for Natural Language Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Samridhi Choudhary, Siegfried Kunzmann, Zheng Zhang, Zi Yang","submitted_at":"2023-06-01T18:32:08Z","abstract_excerpt":"Fine-tuned transformer models have shown superior performances in many natural language tasks. However, the large model size prohibits deploying high-performance transformer models on resource-constrained devices. This paper proposes a quantization-aware tensor-compressed training approach to reduce the model size, arithmetic operations, and ultimately runtime latency of transformer-based models. We compress the embedding and linear layers of transformers into small low-rank tensor cores, which significantly reduces model parameters. A quantization-aware training with learnable scale factors i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.01076","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/2306.01076/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":"2026-07-05T09:12:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GRC10UsSyVm7uDngZzxe2JdZeOq1+4Smtb5sBfYnmn6pnasNJL2/p+BsHL0eDtVlEc0sKxXGC410RUBXuYZxCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T12:40:36.968035Z"},"content_sha256":"58651a8476d2026e156e7901de07499db5e5381836b6fd8b15bbe9fa5e80d7f7","schema_version":"1.0","event_id":"sha256:58651a8476d2026e156e7901de07499db5e5381836b6fd8b15bbe9fa5e80d7f7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NBIWZ3MIKYELQEWCVGJ2FAYYI2/bundle.json","state_url":"https://pith.science/pith/NBIWZ3MIKYELQEWCVGJ2FAYYI2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NBIWZ3MIKYELQEWCVGJ2FAYYI2/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":"2026-08-11T12:40:36Z","links":{"resolver":"https://pith.science/pith/NBIWZ3MIKYELQEWCVGJ2FAYYI2","bundle":"https://pith.science/pith/NBIWZ3MIKYELQEWCVGJ2FAYYI2/bundle.json","state":"https://pith.science/pith/NBIWZ3MIKYELQEWCVGJ2FAYYI2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NBIWZ3MIKYELQEWCVGJ2FAYYI2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:NBIWZ3MIKYELQEWCVGJ2FAYYI2","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":"3c4cf910b5c89c6a7c8c70771588e578281764cb1dc4d3c082591d4d4215e2e3","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-01T18:32:08Z","title_canon_sha256":"912b0613b7a52f7cfe18f6a327193ce5680143db3dc35f7695cd8fd2e9ee4e53"},"schema_version":"1.0","source":{"id":"2306.01076","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.01076","created_at":"2026-07-05T09:12:53Z"},{"alias_kind":"arxiv_version","alias_value":"2306.01076v2","created_at":"2026-07-05T09:12:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.01076","created_at":"2026-07-05T09:12:53Z"},{"alias_kind":"pith_short_12","alias_value":"NBIWZ3MIKYEL","created_at":"2026-07-05T09:12:53Z"},{"alias_kind":"pith_short_16","alias_value":"NBIWZ3MIKYELQEWC","created_at":"2026-07-05T09:12:53Z"},{"alias_kind":"pith_short_8","alias_value":"NBIWZ3MI","created_at":"2026-07-05T09:12:53Z"}],"graph_snapshots":[{"event_id":"sha256:58651a8476d2026e156e7901de07499db5e5381836b6fd8b15bbe9fa5e80d7f7","target":"graph","created_at":"2026-07-05T09:12:53Z","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/2306.01076/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Fine-tuned transformer models have shown superior performances in many natural language tasks. However, the large model size prohibits deploying high-performance transformer models on resource-constrained devices. This paper proposes a quantization-aware tensor-compressed training approach to reduce the model size, arithmetic operations, and ultimately runtime latency of transformer-based models. We compress the embedding and linear layers of transformers into small low-rank tensor cores, which significantly reduces model parameters. A quantization-aware training with learnable scale factors i","authors_text":"Samridhi Choudhary, Siegfried Kunzmann, Zheng Zhang, Zi Yang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-01T18:32:08Z","title":"Quantization-Aware and Tensor-Compressed Training of Transformers for Natural Language Understanding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.01076","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":"sha256:9ec1f787d98dad1d2595e0eb234a4b6240b5a6b3cfe20c42bc25238f95c63d50","target":"record","created_at":"2026-07-05T09:12:53Z","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":"3c4cf910b5c89c6a7c8c70771588e578281764cb1dc4d3c082591d4d4215e2e3","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-01T18:32:08Z","title_canon_sha256":"912b0613b7a52f7cfe18f6a327193ce5680143db3dc35f7695cd8fd2e9ee4e53"},"schema_version":"1.0","source":{"id":"2306.01076","kind":"arxiv","version":2}},"canonical_sha256":"68516ced885608b812c2a993a2831846a64d98335f30cc712194f2ee182b028d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"68516ced885608b812c2a993a2831846a64d98335f30cc712194f2ee182b028d","first_computed_at":"2026-07-05T09:12:53.390092Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:12:53.390092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hscnDMlLapI60jUfhLEmdEkq/IRPQDumAkuVe7VUQxSIM5PECiSpJuptDq+cViyFdWXgKL9IlQUQK5vYZ4NkDw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:12:53.390576Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.01076","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9ec1f787d98dad1d2595e0eb234a4b6240b5a6b3cfe20c42bc25238f95c63d50","sha256:58651a8476d2026e156e7901de07499db5e5381836b6fd8b15bbe9fa5e80d7f7"],"state_sha256":"30626355ba6370e5cfb5562c1092eca109b3f47cc246d8de39e874e8de8b3abf"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XTwDczZa36n1zRu5gBuXnZWwhyMk5t5UN2ak8BM7BPQA01wKmRmQXs2mooYDg0q0LfLyURRR1vIHTGG4GMH3Aw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T12:40:36.973515Z","bundle_sha256":"a3e28a088d4b4f234a2e4500966dbe3150f1999f9c010b8f10812c8991944f9b"}}