{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:BNAHIINSWWTSLMO7IIGYA2N3LN","short_pith_number":"pith:BNAHIINS","canonical_record":{"source":{"id":"2110.04711","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-10T05:44:02Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"278f9cb510f8028b61a67c486510d4762c6e41677e562cbf9824effb852b5a99","abstract_canon_sha256":"cec2d049ddab0af7634ab6edf364f8e827c4ad9abb2de26c56aa38830db8c868"},"schema_version":"1.0"},"canonical_sha256":"0b407421b2b5a725b1df420d8069bb5b76a46d6e67081beb4612c1ea0b530174","source":{"kind":"arxiv","id":"2110.04711","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2110.04711","created_at":"2026-07-05T03:21:20Z"},{"alias_kind":"arxiv_version","alias_value":"2110.04711v1","created_at":"2026-07-05T03:21:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.04711","created_at":"2026-07-05T03:21:20Z"},{"alias_kind":"pith_short_12","alias_value":"BNAHIINSWWTS","created_at":"2026-07-05T03:21:20Z"},{"alias_kind":"pith_short_16","alias_value":"BNAHIINSWWTSLMO7","created_at":"2026-07-05T03:21:20Z"},{"alias_kind":"pith_short_8","alias_value":"BNAHIINS","created_at":"2026-07-05T03:21:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:BNAHIINSWWTSLMO7IIGYA2N3LN","target":"record","payload":{"canonical_record":{"source":{"id":"2110.04711","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-10T05:44:02Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"278f9cb510f8028b61a67c486510d4762c6e41677e562cbf9824effb852b5a99","abstract_canon_sha256":"cec2d049ddab0af7634ab6edf364f8e827c4ad9abb2de26c56aa38830db8c868"},"schema_version":"1.0"},"canonical_sha256":"0b407421b2b5a725b1df420d8069bb5b76a46d6e67081beb4612c1ea0b530174","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:21:20.439893Z","signature_b64":"JJDIE2GG92eYoXzH+tZvysM/1+0KYKLp8lat/v7Mli6819eW8kdyaH/YRnPAR/NEOcxTK+9jQJzA12N6j6WNBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0b407421b2b5a725b1df420d8069bb5b76a46d6e67081beb4612c1ea0b530174","last_reissued_at":"2026-07-05T03:21:20.439466Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:21:20.439466Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2110.04711","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":"2026-07-05T03:21:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"B0rcTnhDHYS6u6QW82Ogm5Wgjxb9YFK0Qs/04RtLA8zWek4Mvy6jY+sgcMR4e4Em0o+LvxUWFwuJL8v7aA8YDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-20T20:31:00.170054Z"},"content_sha256":"7cdb1d18040f6022dfb18716228bf6c9119a07f37dcc5c88a883953ed03bd3e7","schema_version":"1.0","event_id":"sha256:7cdb1d18040f6022dfb18716228bf6c9119a07f37dcc5c88a883953ed03bd3e7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:BNAHIINSWWTSLMO7IIGYA2N3LN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SuperShaper: Task-Agnostic Super Pre-training of BERT Models with Variable Hidden Dimensions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Gowtham Ramesh, Pratyush Kumar, Vinod Ganesan","submitted_at":"2021-10-10T05:44:02Z","abstract_excerpt":"Task-agnostic pre-training followed by task-specific fine-tuning is a default approach to train NLU models. Such models need to be deployed on devices across the cloud and the edge with varying resource and accuracy constraints. For a given task, repeating pre-training and fine-tuning across tens of devices is prohibitively expensive. We propose SuperShaper, a task agnostic pre-training approach which simultaneously pre-trains a large number of Transformer models by varying shapes, i.e., by varying the hidden dimensions across layers. This is enabled by a backbone network with linear bottlenec"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.04711","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/2110.04711/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":"2026-07-05T03:21:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yMYwqWIT+DR2mY5JrNWwhrOsD3wx2mCtQOdPwkYm1qlrzrUryRMS91XmeRx19pHMq8nEk/rEbkMz/qGRgfIpBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-20T20:31:00.170424Z"},"content_sha256":"0a110da0c8140248cb8d2ecb515a6059e998241f51f773dc9f7e5c860b30220c","schema_version":"1.0","event_id":"sha256:0a110da0c8140248cb8d2ecb515a6059e998241f51f773dc9f7e5c860b30220c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BNAHIINSWWTSLMO7IIGYA2N3LN/bundle.json","state_url":"https://pith.science/pith/BNAHIINSWWTSLMO7IIGYA2N3LN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BNAHIINSWWTSLMO7IIGYA2N3LN/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":"2026-07-20T20:31:00Z","links":{"resolver":"https://pith.science/pith/BNAHIINSWWTSLMO7IIGYA2N3LN","bundle":"https://pith.science/pith/BNAHIINSWWTSLMO7IIGYA2N3LN/bundle.json","state":"https://pith.science/pith/BNAHIINSWWTSLMO7IIGYA2N3LN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BNAHIINSWWTSLMO7IIGYA2N3LN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:BNAHIINSWWTSLMO7IIGYA2N3LN","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":"cec2d049ddab0af7634ab6edf364f8e827c4ad9abb2de26c56aa38830db8c868","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-10T05:44:02Z","title_canon_sha256":"278f9cb510f8028b61a67c486510d4762c6e41677e562cbf9824effb852b5a99"},"schema_version":"1.0","source":{"id":"2110.04711","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2110.04711","created_at":"2026-07-05T03:21:20Z"},{"alias_kind":"arxiv_version","alias_value":"2110.04711v1","created_at":"2026-07-05T03:21:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.04711","created_at":"2026-07-05T03:21:20Z"},{"alias_kind":"pith_short_12","alias_value":"BNAHIINSWWTS","created_at":"2026-07-05T03:21:20Z"},{"alias_kind":"pith_short_16","alias_value":"BNAHIINSWWTSLMO7","created_at":"2026-07-05T03:21:20Z"},{"alias_kind":"pith_short_8","alias_value":"BNAHIINS","created_at":"2026-07-05T03:21:20Z"}],"graph_snapshots":[{"event_id":"sha256:0a110da0c8140248cb8d2ecb515a6059e998241f51f773dc9f7e5c860b30220c","target":"graph","created_at":"2026-07-05T03:21:20Z","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/2110.04711/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Task-agnostic pre-training followed by task-specific fine-tuning is a default approach to train NLU models. Such models need to be deployed on devices across the cloud and the edge with varying resource and accuracy constraints. For a given task, repeating pre-training and fine-tuning across tens of devices is prohibitively expensive. We propose SuperShaper, a task agnostic pre-training approach which simultaneously pre-trains a large number of Transformer models by varying shapes, i.e., by varying the hidden dimensions across layers. This is enabled by a backbone network with linear bottlenec","authors_text":"Gowtham Ramesh, Pratyush Kumar, Vinod Ganesan","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-10T05:44:02Z","title":"SuperShaper: Task-Agnostic Super Pre-training of BERT Models with Variable Hidden Dimensions"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.04711","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":"sha256:7cdb1d18040f6022dfb18716228bf6c9119a07f37dcc5c88a883953ed03bd3e7","target":"record","created_at":"2026-07-05T03:21:20Z","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":"cec2d049ddab0af7634ab6edf364f8e827c4ad9abb2de26c56aa38830db8c868","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-10T05:44:02Z","title_canon_sha256":"278f9cb510f8028b61a67c486510d4762c6e41677e562cbf9824effb852b5a99"},"schema_version":"1.0","source":{"id":"2110.04711","kind":"arxiv","version":1}},"canonical_sha256":"0b407421b2b5a725b1df420d8069bb5b76a46d6e67081beb4612c1ea0b530174","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0b407421b2b5a725b1df420d8069bb5b76a46d6e67081beb4612c1ea0b530174","first_computed_at":"2026-07-05T03:21:20.439466Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:21:20.439466Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"JJDIE2GG92eYoXzH+tZvysM/1+0KYKLp8lat/v7Mli6819eW8kdyaH/YRnPAR/NEOcxTK+9jQJzA12N6j6WNBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T03:21:20.439893Z","signed_message":"canonical_sha256_bytes"},"source_id":"2110.04711","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7cdb1d18040f6022dfb18716228bf6c9119a07f37dcc5c88a883953ed03bd3e7","sha256:0a110da0c8140248cb8d2ecb515a6059e998241f51f773dc9f7e5c860b30220c"],"state_sha256":"4638cb409bcc15d7e9a395771ce9aa8de37d2d34126cbe780cf7c4f1b9290209"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AUVgvqOeWX3R8GK3y1jVARGidywj8iSGgcjfR4eZN8jowfyiF843+y6gLFGxp3cLI1cqAXI7uZBKtGj+f4gyDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-20T20:31:00.172551Z","bundle_sha256":"21744b1a6d2b2ac05376da0334f9abdf223aed0dc0ca718e0c9e8e9ba8630017"}}