{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:AK4AI6LH6XE2DQGHNBSANEWBC3","short_pith_number":"pith:AK4AI6LH","canonical_record":{"source":{"id":"2105.13003","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-05-27T08:38:29Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"2061eeb9c923040f9dd21bb68c838686c66c879c5b125279beb0ab680f1aadfc","abstract_canon_sha256":"090f3c3e7f3a4e53ea9153dce119609cf88ac5de555ec3b899c510a5f06f658f"},"schema_version":"1.0"},"canonical_sha256":"02b8047967f5c9a1c0c768640692c116efce7ebc5491246d9bab8da873dd32ad","source":{"kind":"arxiv","id":"2105.13003","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2105.13003","created_at":"2026-07-05T02:43:56Z"},{"alias_kind":"arxiv_version","alias_value":"2105.13003v1","created_at":"2026-07-05T02:43:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.13003","created_at":"2026-07-05T02:43:56Z"},{"alias_kind":"pith_short_12","alias_value":"AK4AI6LH6XE2","created_at":"2026-07-05T02:43:56Z"},{"alias_kind":"pith_short_16","alias_value":"AK4AI6LH6XE2DQGH","created_at":"2026-07-05T02:43:56Z"},{"alias_kind":"pith_short_8","alias_value":"AK4AI6LH","created_at":"2026-07-05T02:43:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:AK4AI6LH6XE2DQGHNBSANEWBC3","target":"record","payload":{"canonical_record":{"source":{"id":"2105.13003","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-05-27T08:38:29Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"2061eeb9c923040f9dd21bb68c838686c66c879c5b125279beb0ab680f1aadfc","abstract_canon_sha256":"090f3c3e7f3a4e53ea9153dce119609cf88ac5de555ec3b899c510a5f06f658f"},"schema_version":"1.0"},"canonical_sha256":"02b8047967f5c9a1c0c768640692c116efce7ebc5491246d9bab8da873dd32ad","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:43:56.318353Z","signature_b64":"KnGPfe3OBB3NWEpp8Hz9/2ZDsQrsTSFcmKU/sqxayKJnxgTu5dEJ1qjvo4cSXkOb/xmBnQK30H4qWLOu6WSBDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"02b8047967f5c9a1c0c768640692c116efce7ebc5491246d9bab8da873dd32ad","last_reissued_at":"2026-07-05T02:43:56.317605Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:43:56.317605Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2105.13003","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":"2026-07-05T02:43:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HuEcTxTe6lsQa64KtT74h/qW9HUeidbZnphKiux8sK1rCCSI7OPBuWAeYFMl9vsi68PEsEyxjiuJBIZZzRmnCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T14:33:53.991975Z"},"content_sha256":"236d9e1d2d4d82d3a6574e52ba71e04b9981eae45dd6320fc9bc95e091573c33","schema_version":"1.0","event_id":"sha256:236d9e1d2d4d82d3a6574e52ba71e04b9981eae45dd6320fc9bc95e091573c33"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:AK4AI6LH6XE2DQGHNBSANEWBC3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Rethinking InfoNCE: How Many Negative Samples Do You Need?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.LG","authors_text":"Chuhan Wu, Fangzhao Wu, Yongfeng Huang","submitted_at":"2021-05-27T08:38:29Z","abstract_excerpt":"InfoNCE loss is a widely used loss function for contrastive model training. It aims to estimate the mutual information between a pair of variables by discriminating between each positive pair and its associated $K$ negative pairs. It is proved that when the sample labels are clean, the lower bound of mutual information estimation is tighter when more negative samples are incorporated, which usually yields better model performance. However, in many real-world tasks the labels often contain noise, and incorporating too many noisy negative samples for model training may be suboptimal. In this pap"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.13003","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/2105.13003/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":"2026-07-05T02:43:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F7v2sVBEUqXcsPOg3okdy3cJXPkJW6Cf4SkHEWtFEMbNgCAyASKodoiPacwdITjgWO0oGH/2L/5lCHS+7hbaDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T14:33:53.992479Z"},"content_sha256":"bc9cc9cc776039b2ec48fa836dbe3896cb6e82565578b222d5c5c795e34f2f64","schema_version":"1.0","event_id":"sha256:bc9cc9cc776039b2ec48fa836dbe3896cb6e82565578b222d5c5c795e34f2f64"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AK4AI6LH6XE2DQGHNBSANEWBC3/bundle.json","state_url":"https://pith.science/pith/AK4AI6LH6XE2DQGHNBSANEWBC3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AK4AI6LH6XE2DQGHNBSANEWBC3/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":"2026-08-09T14:33:53Z","links":{"resolver":"https://pith.science/pith/AK4AI6LH6XE2DQGHNBSANEWBC3","bundle":"https://pith.science/pith/AK4AI6LH6XE2DQGHNBSANEWBC3/bundle.json","state":"https://pith.science/pith/AK4AI6LH6XE2DQGHNBSANEWBC3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AK4AI6LH6XE2DQGHNBSANEWBC3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:AK4AI6LH6XE2DQGHNBSANEWBC3","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":"090f3c3e7f3a4e53ea9153dce119609cf88ac5de555ec3b899c510a5f06f658f","cross_cats_sorted":["cs.IR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-05-27T08:38:29Z","title_canon_sha256":"2061eeb9c923040f9dd21bb68c838686c66c879c5b125279beb0ab680f1aadfc"},"schema_version":"1.0","source":{"id":"2105.13003","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2105.13003","created_at":"2026-07-05T02:43:56Z"},{"alias_kind":"arxiv_version","alias_value":"2105.13003v1","created_at":"2026-07-05T02:43:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.13003","created_at":"2026-07-05T02:43:56Z"},{"alias_kind":"pith_short_12","alias_value":"AK4AI6LH6XE2","created_at":"2026-07-05T02:43:56Z"},{"alias_kind":"pith_short_16","alias_value":"AK4AI6LH6XE2DQGH","created_at":"2026-07-05T02:43:56Z"},{"alias_kind":"pith_short_8","alias_value":"AK4AI6LH","created_at":"2026-07-05T02:43:56Z"}],"graph_snapshots":[{"event_id":"sha256:bc9cc9cc776039b2ec48fa836dbe3896cb6e82565578b222d5c5c795e34f2f64","target":"graph","created_at":"2026-07-05T02:43:56Z","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/2105.13003/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"InfoNCE loss is a widely used loss function for contrastive model training. It aims to estimate the mutual information between a pair of variables by discriminating between each positive pair and its associated $K$ negative pairs. It is proved that when the sample labels are clean, the lower bound of mutual information estimation is tighter when more negative samples are incorporated, which usually yields better model performance. However, in many real-world tasks the labels often contain noise, and incorporating too many noisy negative samples for model training may be suboptimal. In this pap","authors_text":"Chuhan Wu, Fangzhao Wu, Yongfeng Huang","cross_cats":["cs.IR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-05-27T08:38:29Z","title":"Rethinking InfoNCE: How Many Negative Samples Do You Need?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.13003","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":"sha256:236d9e1d2d4d82d3a6574e52ba71e04b9981eae45dd6320fc9bc95e091573c33","target":"record","created_at":"2026-07-05T02:43:56Z","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":"090f3c3e7f3a4e53ea9153dce119609cf88ac5de555ec3b899c510a5f06f658f","cross_cats_sorted":["cs.IR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-05-27T08:38:29Z","title_canon_sha256":"2061eeb9c923040f9dd21bb68c838686c66c879c5b125279beb0ab680f1aadfc"},"schema_version":"1.0","source":{"id":"2105.13003","kind":"arxiv","version":1}},"canonical_sha256":"02b8047967f5c9a1c0c768640692c116efce7ebc5491246d9bab8da873dd32ad","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"02b8047967f5c9a1c0c768640692c116efce7ebc5491246d9bab8da873dd32ad","first_computed_at":"2026-07-05T02:43:56.317605Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:43:56.317605Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KnGPfe3OBB3NWEpp8Hz9/2ZDsQrsTSFcmKU/sqxayKJnxgTu5dEJ1qjvo4cSXkOb/xmBnQK30H4qWLOu6WSBDw==","signature_status":"signed_v1","signed_at":"2026-07-05T02:43:56.318353Z","signed_message":"canonical_sha256_bytes"},"source_id":"2105.13003","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:236d9e1d2d4d82d3a6574e52ba71e04b9981eae45dd6320fc9bc95e091573c33","sha256:bc9cc9cc776039b2ec48fa836dbe3896cb6e82565578b222d5c5c795e34f2f64"],"state_sha256":"57e500464756db86761ae049b9a9a43408efc9ed371855f4e30215bec93b999f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rRfBmIAiUa6zWqAkfcbSFKjDjm8eeh6teQw+WQXGbxBcehUUzqtIX5I4GwBWK1aLsdCVQ77DguyJ1LnpXCOGBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T14:33:53.997337Z","bundle_sha256":"461dfc06f66051faadefc77684c6018cb85e21515bb9b0075b587804258aa602"}}