{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:ATVTDHCOE54R5VS6UGCBS7XVOM","short_pith_number":"pith:ATVTDHCO","canonical_record":{"source":{"id":"2212.08853","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-17T11:56:21Z","cross_cats_sorted":[],"title_canon_sha256":"2f3c9f59caaa025c90ff0249ee33da89a7f3d81d2cb2cd7d76a9da6414552b47","abstract_canon_sha256":"4b894a1cabedf3c6e7c8cef0f7c81656297e5b75da3acc6374d2e35ef4b6095f"},"schema_version":"1.0"},"canonical_sha256":"04eb319c4e27791ed65ea184197ef573078ddc375e3f3484284006fcbb82d5c8","source":{"kind":"arxiv","id":"2212.08853","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2212.08853","created_at":"2026-07-05T06:09:07Z"},{"alias_kind":"arxiv_version","alias_value":"2212.08853v2","created_at":"2026-07-05T06:09:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.08853","created_at":"2026-07-05T06:09:07Z"},{"alias_kind":"pith_short_12","alias_value":"ATVTDHCOE54R","created_at":"2026-07-05T06:09:07Z"},{"alias_kind":"pith_short_16","alias_value":"ATVTDHCOE54R5VS6","created_at":"2026-07-05T06:09:07Z"},{"alias_kind":"pith_short_8","alias_value":"ATVTDHCO","created_at":"2026-07-05T06:09:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:ATVTDHCOE54R5VS6UGCBS7XVOM","target":"record","payload":{"canonical_record":{"source":{"id":"2212.08853","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-17T11:56:21Z","cross_cats_sorted":[],"title_canon_sha256":"2f3c9f59caaa025c90ff0249ee33da89a7f3d81d2cb2cd7d76a9da6414552b47","abstract_canon_sha256":"4b894a1cabedf3c6e7c8cef0f7c81656297e5b75da3acc6374d2e35ef4b6095f"},"schema_version":"1.0"},"canonical_sha256":"04eb319c4e27791ed65ea184197ef573078ddc375e3f3484284006fcbb82d5c8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:09:07.472166Z","signature_b64":"k+oSQgt+4AsUxYGdfhrXX5zobydq5YjKSVdokfAubXPSFZGt2nM/sBm8tvmoB2QwJ4KgzVV8JEarP4OcZVbKAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"04eb319c4e27791ed65ea184197ef573078ddc375e3f3484284006fcbb82d5c8","last_reissued_at":"2026-07-05T06:09:07.471737Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:09:07.471737Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2212.08853","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":"2026-07-05T06:09:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GYObx8lNktk6TpOZaMMcgKvAqspncrFirPXomsT0UW40qpLEiuMKxbn4Z5CkDRi18suok+pLuvNvmj1R91Z5Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T09:25:43.636003Z"},"content_sha256":"a6cab8a1641afada548877778ed8e73170dc50691986adea1092c82e58663499","schema_version":"1.0","event_id":"sha256:a6cab8a1641afada548877778ed8e73170dc50691986adea1092c82e58663499"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:ATVTDHCOE54R5VS6UGCBS7XVOM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"HyPe: Better Pre-trained Language Model Fine-tuning with Hidden Representation Perturbation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chuanqi Tan, Fei Huang, Hongyi Yuan, Songfang Huang, Zheng Yuan","submitted_at":"2022-12-17T11:56:21Z","abstract_excerpt":"Language models with the Transformers structure have shown great performance in natural language processing. However, there still poses problems when fine-tuning pre-trained language models on downstream tasks, such as over-fitting or representation collapse. In this work, we propose HyPe, a simple yet effective fine-tuning technique to alleviate such problems by perturbing hidden representations of Transformers layers. Unlike previous works that only add noise to inputs or parameters, we argue that the hidden representations of Transformers layers convey more diverse and meaningful language i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.08853","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/2212.08853/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":"2026-07-05T06:09:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WBhngXGTk9CUFHAU+hJ2yuSNRKbpEsgOltZEws1LVg/7SUxp2Y3h3r9TIox/CBMAdCftAy25hKDXz0XAPb2LAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T09:25:43.636518Z"},"content_sha256":"83883d9de98e10094cae866916e009955965f5d28221f45a26e8f3130869d2c8","schema_version":"1.0","event_id":"sha256:83883d9de98e10094cae866916e009955965f5d28221f45a26e8f3130869d2c8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ATVTDHCOE54R5VS6UGCBS7XVOM/bundle.json","state_url":"https://pith.science/pith/ATVTDHCOE54R5VS6UGCBS7XVOM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ATVTDHCOE54R5VS6UGCBS7XVOM/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":"2026-08-07T09:25:43Z","links":{"resolver":"https://pith.science/pith/ATVTDHCOE54R5VS6UGCBS7XVOM","bundle":"https://pith.science/pith/ATVTDHCOE54R5VS6UGCBS7XVOM/bundle.json","state":"https://pith.science/pith/ATVTDHCOE54R5VS6UGCBS7XVOM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ATVTDHCOE54R5VS6UGCBS7XVOM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:ATVTDHCOE54R5VS6UGCBS7XVOM","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":"4b894a1cabedf3c6e7c8cef0f7c81656297e5b75da3acc6374d2e35ef4b6095f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-17T11:56:21Z","title_canon_sha256":"2f3c9f59caaa025c90ff0249ee33da89a7f3d81d2cb2cd7d76a9da6414552b47"},"schema_version":"1.0","source":{"id":"2212.08853","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2212.08853","created_at":"2026-07-05T06:09:07Z"},{"alias_kind":"arxiv_version","alias_value":"2212.08853v2","created_at":"2026-07-05T06:09:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.08853","created_at":"2026-07-05T06:09:07Z"},{"alias_kind":"pith_short_12","alias_value":"ATVTDHCOE54R","created_at":"2026-07-05T06:09:07Z"},{"alias_kind":"pith_short_16","alias_value":"ATVTDHCOE54R5VS6","created_at":"2026-07-05T06:09:07Z"},{"alias_kind":"pith_short_8","alias_value":"ATVTDHCO","created_at":"2026-07-05T06:09:07Z"}],"graph_snapshots":[{"event_id":"sha256:83883d9de98e10094cae866916e009955965f5d28221f45a26e8f3130869d2c8","target":"graph","created_at":"2026-07-05T06:09:07Z","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/2212.08853/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language models with the Transformers structure have shown great performance in natural language processing. However, there still poses problems when fine-tuning pre-trained language models on downstream tasks, such as over-fitting or representation collapse. In this work, we propose HyPe, a simple yet effective fine-tuning technique to alleviate such problems by perturbing hidden representations of Transformers layers. Unlike previous works that only add noise to inputs or parameters, we argue that the hidden representations of Transformers layers convey more diverse and meaningful language i","authors_text":"Chuanqi Tan, Fei Huang, Hongyi Yuan, Songfang Huang, Zheng Yuan","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-17T11:56:21Z","title":"HyPe: Better Pre-trained Language Model Fine-tuning with Hidden Representation Perturbation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.08853","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:a6cab8a1641afada548877778ed8e73170dc50691986adea1092c82e58663499","target":"record","created_at":"2026-07-05T06:09:07Z","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":"4b894a1cabedf3c6e7c8cef0f7c81656297e5b75da3acc6374d2e35ef4b6095f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-17T11:56:21Z","title_canon_sha256":"2f3c9f59caaa025c90ff0249ee33da89a7f3d81d2cb2cd7d76a9da6414552b47"},"schema_version":"1.0","source":{"id":"2212.08853","kind":"arxiv","version":2}},"canonical_sha256":"04eb319c4e27791ed65ea184197ef573078ddc375e3f3484284006fcbb82d5c8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"04eb319c4e27791ed65ea184197ef573078ddc375e3f3484284006fcbb82d5c8","first_computed_at":"2026-07-05T06:09:07.471737Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:09:07.471737Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"k+oSQgt+4AsUxYGdfhrXX5zobydq5YjKSVdokfAubXPSFZGt2nM/sBm8tvmoB2QwJ4KgzVV8JEarP4OcZVbKAw==","signature_status":"signed_v1","signed_at":"2026-07-05T06:09:07.472166Z","signed_message":"canonical_sha256_bytes"},"source_id":"2212.08853","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a6cab8a1641afada548877778ed8e73170dc50691986adea1092c82e58663499","sha256:83883d9de98e10094cae866916e009955965f5d28221f45a26e8f3130869d2c8"],"state_sha256":"53b75f156ebf24d5223c0e8a97b1e6cc61ca3b7dbab2ab877b016e953c14a808"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"av9ocVpOOH6Bk71ZLDIGhcqx3hw1IHoN+refe+fi6PsZ0NyYiEyLrDPSftIqYDKAkYIfmuy4XFtxdSETx4OvDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T09:25:43.640473Z","bundle_sha256":"659318f85ec4eaac208ad2fd5b0428305ef21f6fd6894c0b489355a5bb325d11"}}