{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:55PFJKS3WC36KVKSLFTJUWIKEQ","short_pith_number":"pith:55PFJKS3","canonical_record":{"source":{"id":"2112.12650","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-12-23T15:37:58Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"24fc4152f8cb7815544d3ed6e0405d99eaf84492ddbb971f79cc2065e956c100","abstract_canon_sha256":"4809168501dec5e03ffdd1bb55445ebdc64087a17c4e9bf96a9f9bc3944778e7"},"schema_version":"1.0"},"canonical_sha256":"ef5e54aa5bb0b7e5555259669a590a242e8b5e9f19b864c65ba1a5b8bee20ecc","source":{"kind":"arxiv","id":"2112.12650","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2112.12650","created_at":"2026-07-05T04:14:27Z"},{"alias_kind":"arxiv_version","alias_value":"2112.12650v3","created_at":"2026-07-05T04:14:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.12650","created_at":"2026-07-05T04:14:27Z"},{"alias_kind":"pith_short_12","alias_value":"55PFJKS3WC36","created_at":"2026-07-05T04:14:27Z"},{"alias_kind":"pith_short_16","alias_value":"55PFJKS3WC36KVKS","created_at":"2026-07-05T04:14:27Z"},{"alias_kind":"pith_short_8","alias_value":"55PFJKS3","created_at":"2026-07-05T04:14:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:55PFJKS3WC36KVKSLFTJUWIKEQ","target":"record","payload":{"canonical_record":{"source":{"id":"2112.12650","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-12-23T15:37:58Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"24fc4152f8cb7815544d3ed6e0405d99eaf84492ddbb971f79cc2065e956c100","abstract_canon_sha256":"4809168501dec5e03ffdd1bb55445ebdc64087a17c4e9bf96a9f9bc3944778e7"},"schema_version":"1.0"},"canonical_sha256":"ef5e54aa5bb0b7e5555259669a590a242e8b5e9f19b864c65ba1a5b8bee20ecc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:14:27.243168Z","signature_b64":"9773xCQIKiRRdxF1ktKMpnEphrrA8ejpsPvriV2s7LxLWNSY7trc0NBMaMCzcOw5BOdOfd74CQ+gni1rTpXfCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ef5e54aa5bb0b7e5555259669a590a242e8b5e9f19b864c65ba1a5b8bee20ecc","last_reissued_at":"2026-07-05T04:14:27.242735Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:14:27.242735Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2112.12650","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:14:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tjNqLCf1SsccNiV3cmvkxnoLX43Ap6DA0C/hskMy9TgXAIbYpbk1KtOL+nmwW5j5J9C3jNqp+g1sFeJNb8JhAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-26T19:46:37.536957Z"},"content_sha256":"ae10479c2506206f99f4c9b8791c3399caccfde60fda8dd0a8583c42dbd3bbef","schema_version":"1.0","event_id":"sha256:ae10479c2506206f99f4c9b8791c3399caccfde60fda8dd0a8583c42dbd3bbef"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:55PFJKS3WC36KVKSLFTJUWIKEQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Distilling the Knowledge of Romanian BERTs Using Multiple Teachers","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Andrei-Marius Avram, Dan Tufi\\c{s}, Darius Catrina, Dumitru-Clementin Cercel, Mihai Dasc\\u{a}lu, Traian Rebedea, Vasile P\\u{a}i\\c{s}","submitted_at":"2021-12-23T15:37:58Z","abstract_excerpt":"Running large-scale pre-trained language models in computationally constrained environments remains a challenging problem yet to be addressed, while transfer learning from these models has become prevalent in Natural Language Processing tasks. Several solutions, including knowledge distillation, network quantization, or network pruning have been previously proposed; however, these approaches focus mostly on the English language, thus widening the gap when considering low-resource languages. In this work, we introduce three light and fast versions of distilled BERT models for the Romanian langu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.12650","kind":"arxiv","version":3},"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/2112.12650/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":"2026-07-05T04:14:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nOPVGEEj4gDBJGifpafLs0OO9bE9KVIq1IGaTj7kYGKEFFzG3SK2NTjgpIy85rVThfrk6JDnSXEFeIA5X3m0Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-26T19:46:37.537334Z"},"content_sha256":"2ab4e58de963aa570e56be5747f0a0dadfd9a321325ddd37876a232b84e40665","schema_version":"1.0","event_id":"sha256:2ab4e58de963aa570e56be5747f0a0dadfd9a321325ddd37876a232b84e40665"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/55PFJKS3WC36KVKSLFTJUWIKEQ/bundle.json","state_url":"https://pith.science/pith/55PFJKS3WC36KVKSLFTJUWIKEQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/55PFJKS3WC36KVKSLFTJUWIKEQ/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":"2026-07-26T19:46:37Z","links":{"resolver":"https://pith.science/pith/55PFJKS3WC36KVKSLFTJUWIKEQ","bundle":"https://pith.science/pith/55PFJKS3WC36KVKSLFTJUWIKEQ/bundle.json","state":"https://pith.science/pith/55PFJKS3WC36KVKSLFTJUWIKEQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/55PFJKS3WC36KVKSLFTJUWIKEQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:55PFJKS3WC36KVKSLFTJUWIKEQ","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":"4809168501dec5e03ffdd1bb55445ebdc64087a17c4e9bf96a9f9bc3944778e7","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-12-23T15:37:58Z","title_canon_sha256":"24fc4152f8cb7815544d3ed6e0405d99eaf84492ddbb971f79cc2065e956c100"},"schema_version":"1.0","source":{"id":"2112.12650","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2112.12650","created_at":"2026-07-05T04:14:27Z"},{"alias_kind":"arxiv_version","alias_value":"2112.12650v3","created_at":"2026-07-05T04:14:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.12650","created_at":"2026-07-05T04:14:27Z"},{"alias_kind":"pith_short_12","alias_value":"55PFJKS3WC36","created_at":"2026-07-05T04:14:27Z"},{"alias_kind":"pith_short_16","alias_value":"55PFJKS3WC36KVKS","created_at":"2026-07-05T04:14:27Z"},{"alias_kind":"pith_short_8","alias_value":"55PFJKS3","created_at":"2026-07-05T04:14:27Z"}],"graph_snapshots":[{"event_id":"sha256:2ab4e58de963aa570e56be5747f0a0dadfd9a321325ddd37876a232b84e40665","target":"graph","created_at":"2026-07-05T04:14:27Z","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/2112.12650/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Running large-scale pre-trained language models in computationally constrained environments remains a challenging problem yet to be addressed, while transfer learning from these models has become prevalent in Natural Language Processing tasks. Several solutions, including knowledge distillation, network quantization, or network pruning have been previously proposed; however, these approaches focus mostly on the English language, thus widening the gap when considering low-resource languages. In this work, we introduce three light and fast versions of distilled BERT models for the Romanian langu","authors_text":"Andrei-Marius Avram, Dan Tufi\\c{s}, Darius Catrina, Dumitru-Clementin Cercel, Mihai Dasc\\u{a}lu, Traian Rebedea, Vasile P\\u{a}i\\c{s}","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-12-23T15:37:58Z","title":"Distilling the Knowledge of Romanian BERTs Using Multiple Teachers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.12650","kind":"arxiv","version":3},"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:ae10479c2506206f99f4c9b8791c3399caccfde60fda8dd0a8583c42dbd3bbef","target":"record","created_at":"2026-07-05T04:14:27Z","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":"4809168501dec5e03ffdd1bb55445ebdc64087a17c4e9bf96a9f9bc3944778e7","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-12-23T15:37:58Z","title_canon_sha256":"24fc4152f8cb7815544d3ed6e0405d99eaf84492ddbb971f79cc2065e956c100"},"schema_version":"1.0","source":{"id":"2112.12650","kind":"arxiv","version":3}},"canonical_sha256":"ef5e54aa5bb0b7e5555259669a590a242e8b5e9f19b864c65ba1a5b8bee20ecc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ef5e54aa5bb0b7e5555259669a590a242e8b5e9f19b864c65ba1a5b8bee20ecc","first_computed_at":"2026-07-05T04:14:27.242735Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:14:27.242735Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9773xCQIKiRRdxF1ktKMpnEphrrA8ejpsPvriV2s7LxLWNSY7trc0NBMaMCzcOw5BOdOfd74CQ+gni1rTpXfCA==","signature_status":"signed_v1","signed_at":"2026-07-05T04:14:27.243168Z","signed_message":"canonical_sha256_bytes"},"source_id":"2112.12650","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ae10479c2506206f99f4c9b8791c3399caccfde60fda8dd0a8583c42dbd3bbef","sha256:2ab4e58de963aa570e56be5747f0a0dadfd9a321325ddd37876a232b84e40665"],"state_sha256":"59f02b041a9605149f83aa542ed5b50e04b1d2e08c7204a025b59e29a9a9140b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0/16B5s+S6MZXA5bregCCBSF2bN8sWMUZ/PSK9mFcq9a9IfKNyH/ZbgeFVyvzSRZmpUveFCEnAYlQT+Wc0ZWAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-26T19:46:37.539761Z","bundle_sha256":"a8f05d709e59b9ed39e10702f3a84387f8cbe7f544d280509351e2000010ebd6"}}