{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2016:FEZU2V3US2J5WCEZSBDYPXXN5Y","short_pith_number":"pith:FEZU2V3U","canonical_record":{"source":{"id":"1607.00091","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.AP","submitted_at":"2016-07-01T01:10:35Z","cross_cats_sorted":[],"title_canon_sha256":"3de7cd6a55f20a9157cc7e11507109768126112c7efe8f1bb64f16dcb9528cb8","abstract_canon_sha256":"938eeee1348fcb096d8c4ed55605a05ca88b692320f1ea0b4eab27e937b5709b"},"schema_version":"1.0"},"canonical_sha256":"29334d57749693db0899904787deedee0eaf3e2b23ebaf06c3973b55a50ede88","source":{"kind":"arxiv","id":"1607.00091","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1607.00091","created_at":"2026-05-18T01:11:38Z"},{"alias_kind":"arxiv_version","alias_value":"1607.00091v1","created_at":"2026-05-18T01:11:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1607.00091","created_at":"2026-05-18T01:11:38Z"},{"alias_kind":"pith_short_12","alias_value":"FEZU2V3US2J5","created_at":"2026-05-18T12:30:15Z"},{"alias_kind":"pith_short_16","alias_value":"FEZU2V3US2J5WCEZ","created_at":"2026-05-18T12:30:15Z"},{"alias_kind":"pith_short_8","alias_value":"FEZU2V3U","created_at":"2026-05-18T12:30:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2016:FEZU2V3US2J5WCEZSBDYPXXN5Y","target":"record","payload":{"canonical_record":{"source":{"id":"1607.00091","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.AP","submitted_at":"2016-07-01T01:10:35Z","cross_cats_sorted":[],"title_canon_sha256":"3de7cd6a55f20a9157cc7e11507109768126112c7efe8f1bb64f16dcb9528cb8","abstract_canon_sha256":"938eeee1348fcb096d8c4ed55605a05ca88b692320f1ea0b4eab27e937b5709b"},"schema_version":"1.0"},"canonical_sha256":"29334d57749693db0899904787deedee0eaf3e2b23ebaf06c3973b55a50ede88","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:11:38.075622Z","signature_b64":"wmVegm/Nmwq5Gej9TTM/C4f1JAciXPmu00MYYQ+dD5L5f90fv6Bt5P5vb88WTwC5EX7TGyi5SgnRI25I5PHrDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"29334d57749693db0899904787deedee0eaf3e2b23ebaf06c3973b55a50ede88","last_reissued_at":"2026-05-18T01:11:38.075254Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:11:38.075254Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1607.00091","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":"2026-05-18T01:11:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GhMeR1DeiAxtaFAoDXKQ+kDQFcErX+76OrtbDr2aXZzolDtGtDR0zLhbQOHmo5Oae3t00JRIoXGP+yexa9G2Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T10:38:30.897430Z"},"content_sha256":"cf79879bc474d96b19dfe3bc09bb556acf1dc02747afe6427e5cac289b8065cd","schema_version":"1.0","event_id":"sha256:cf79879bc474d96b19dfe3bc09bb556acf1dc02747afe6427e5cac289b8065cd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2016:FEZU2V3US2J5WCEZSBDYPXXN5Y","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reducing overfitting in challenge-based competitions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"stat.AP","authors_text":"Andrew D Trister, Brian M Bot, Bruce R Hoff, Chris Bare, Elias Chaibub Neto, Gustavo Stolovitzky, James C Costello, Julio Saez-Rodrigues, Justin Guinney, Lara Magravite, Pablo Meyer, Thea Norman, Thomas Yu","submitted_at":"2016-07-01T01:10:35Z","abstract_excerpt":"Over-fitting is a dreaded foe in challenge-based competitions. Because participants rely on public leaderboards to evaluate and refine their models, there is always the danger they might over-fit to the holdout data supporting the leaderboard. The recently published Ladder algorithm aims to address this problem by preventing the participants from exploiting willingly or inadvertently minor fluctuations in public leaderboard scores during model refinement. In this paper, we report a vulnerability of the Ladder that induces severe over-fitting of the leaderboard when the sample size is small. To"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1607.00091","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":""},"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":"2026-05-18T01:11:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"73judpRYQ4PbSVNBUEoOjdK81eIaxqM2vDYxwC9vt1oqXG+P3q+XRkcZyvi8OV6DgiMQSfsqfFjyyZovJPhlBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T10:38:30.898369Z"},"content_sha256":"9016309f524c1c4dadc6968cc1d09c80b7885cbd5ecb2e5f14c174bc9188512d","schema_version":"1.0","event_id":"sha256:9016309f524c1c4dadc6968cc1d09c80b7885cbd5ecb2e5f14c174bc9188512d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FEZU2V3US2J5WCEZSBDYPXXN5Y/bundle.json","state_url":"https://pith.science/pith/FEZU2V3US2J5WCEZSBDYPXXN5Y/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FEZU2V3US2J5WCEZSBDYPXXN5Y/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":"2026-08-18T10:38:30Z","links":{"resolver":"https://pith.science/pith/FEZU2V3US2J5WCEZSBDYPXXN5Y","bundle":"https://pith.science/pith/FEZU2V3US2J5WCEZSBDYPXXN5Y/bundle.json","state":"https://pith.science/pith/FEZU2V3US2J5WCEZSBDYPXXN5Y/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FEZU2V3US2J5WCEZSBDYPXXN5Y/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2016:FEZU2V3US2J5WCEZSBDYPXXN5Y","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":"938eeee1348fcb096d8c4ed55605a05ca88b692320f1ea0b4eab27e937b5709b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.AP","submitted_at":"2016-07-01T01:10:35Z","title_canon_sha256":"3de7cd6a55f20a9157cc7e11507109768126112c7efe8f1bb64f16dcb9528cb8"},"schema_version":"1.0","source":{"id":"1607.00091","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1607.00091","created_at":"2026-05-18T01:11:38Z"},{"alias_kind":"arxiv_version","alias_value":"1607.00091v1","created_at":"2026-05-18T01:11:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1607.00091","created_at":"2026-05-18T01:11:38Z"},{"alias_kind":"pith_short_12","alias_value":"FEZU2V3US2J5","created_at":"2026-05-18T12:30:15Z"},{"alias_kind":"pith_short_16","alias_value":"FEZU2V3US2J5WCEZ","created_at":"2026-05-18T12:30:15Z"},{"alias_kind":"pith_short_8","alias_value":"FEZU2V3U","created_at":"2026-05-18T12:30:15Z"}],"graph_snapshots":[{"event_id":"sha256:9016309f524c1c4dadc6968cc1d09c80b7885cbd5ecb2e5f14c174bc9188512d","target":"graph","created_at":"2026-05-18T01:11:38Z","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"},"paper":{"abstract_excerpt":"Over-fitting is a dreaded foe in challenge-based competitions. Because participants rely on public leaderboards to evaluate and refine their models, there is always the danger they might over-fit to the holdout data supporting the leaderboard. The recently published Ladder algorithm aims to address this problem by preventing the participants from exploiting willingly or inadvertently minor fluctuations in public leaderboard scores during model refinement. In this paper, we report a vulnerability of the Ladder that induces severe over-fitting of the leaderboard when the sample size is small. To","authors_text":"Andrew D Trister, Brian M Bot, Bruce R Hoff, Chris Bare, Elias Chaibub Neto, Gustavo Stolovitzky, James C Costello, Julio Saez-Rodrigues, Justin Guinney, Lara Magravite, Pablo Meyer, Thea Norman, Thomas Yu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.AP","submitted_at":"2016-07-01T01:10:35Z","title":"Reducing overfitting in challenge-based competitions"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1607.00091","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:cf79879bc474d96b19dfe3bc09bb556acf1dc02747afe6427e5cac289b8065cd","target":"record","created_at":"2026-05-18T01:11:38Z","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":"938eeee1348fcb096d8c4ed55605a05ca88b692320f1ea0b4eab27e937b5709b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.AP","submitted_at":"2016-07-01T01:10:35Z","title_canon_sha256":"3de7cd6a55f20a9157cc7e11507109768126112c7efe8f1bb64f16dcb9528cb8"},"schema_version":"1.0","source":{"id":"1607.00091","kind":"arxiv","version":1}},"canonical_sha256":"29334d57749693db0899904787deedee0eaf3e2b23ebaf06c3973b55a50ede88","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"29334d57749693db0899904787deedee0eaf3e2b23ebaf06c3973b55a50ede88","first_computed_at":"2026-05-18T01:11:38.075254Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T01:11:38.075254Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wmVegm/Nmwq5Gej9TTM/C4f1JAciXPmu00MYYQ+dD5L5f90fv6Bt5P5vb88WTwC5EX7TGyi5SgnRI25I5PHrDw==","signature_status":"signed_v1","signed_at":"2026-05-18T01:11:38.075622Z","signed_message":"canonical_sha256_bytes"},"source_id":"1607.00091","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cf79879bc474d96b19dfe3bc09bb556acf1dc02747afe6427e5cac289b8065cd","sha256:9016309f524c1c4dadc6968cc1d09c80b7885cbd5ecb2e5f14c174bc9188512d"],"state_sha256":"089c0ef863f28f7d3dbae4b5157700d41fac477c4f97c21ee62c4dabebebfee6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+uMBsWF9iFFAKAa3orH/Vv+LsO2CJZLFQK2OC/f7CHGI5dCmEpKJhlH6WyoWMkt8ARdaOCJQsVllfmar+y8BCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T10:38:30.906358Z","bundle_sha256":"f20ea0f0854429c0336c5d7fdaecec8d8c1de617ed04cd65cdef6a5f20805286"}}