{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:KHLIKA4AKUQDTZRUCJC2ZHENVP","short_pith_number":"pith:KHLIKA4A","canonical_record":{"source":{"id":"1707.03073","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-07-10T22:01:44Z","cross_cats_sorted":[],"title_canon_sha256":"b3efe2bd834bb945fbc794b605ebb957d075d8151b6ae8a391d11a234c843934","abstract_canon_sha256":"d56f9c466121fd17b51b4ac49934f3fa8c76eb81a2e1569b7872bd4381efbbf5"},"schema_version":"1.0"},"canonical_sha256":"51d6850380552039e6341245ac9c8dabc2b434a820158072cb6cbb507915d06f","source":{"kind":"arxiv","id":"1707.03073","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1707.03073","created_at":"2026-05-18T00:40:18Z"},{"alias_kind":"arxiv_version","alias_value":"1707.03073v2","created_at":"2026-05-18T00:40:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1707.03073","created_at":"2026-05-18T00:40:18Z"},{"alias_kind":"pith_short_12","alias_value":"KHLIKA4AKUQD","created_at":"2026-05-18T12:31:24Z"},{"alias_kind":"pith_short_16","alias_value":"KHLIKA4AKUQDTZRU","created_at":"2026-05-18T12:31:24Z"},{"alias_kind":"pith_short_8","alias_value":"KHLIKA4A","created_at":"2026-05-18T12:31:24Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:KHLIKA4AKUQDTZRUCJC2ZHENVP","target":"record","payload":{"canonical_record":{"source":{"id":"1707.03073","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-07-10T22:01:44Z","cross_cats_sorted":[],"title_canon_sha256":"b3efe2bd834bb945fbc794b605ebb957d075d8151b6ae8a391d11a234c843934","abstract_canon_sha256":"d56f9c466121fd17b51b4ac49934f3fa8c76eb81a2e1569b7872bd4381efbbf5"},"schema_version":"1.0"},"canonical_sha256":"51d6850380552039e6341245ac9c8dabc2b434a820158072cb6cbb507915d06f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:40:18.886531Z","signature_b64":"ux/3pHS+dsrodzg7baC9YyzZyQgw+NPCX69aTN6/aJkVVx/0Rxx76gIMIKYGExSEI4g6AgyKhTl6TJDe+yhCCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"51d6850380552039e6341245ac9c8dabc2b434a820158072cb6cbb507915d06f","last_reissued_at":"2026-05-18T00:40:18.885877Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:40:18.885877Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1707.03073","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":"2026-05-18T00:40:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WmHXQFC5KYmdZiJpkGN/VLt92BmZSbhk4rRWdJkgQcG8m5SEqMnqpE/a/NJPvx22w6iQmFh5nVM77vNb6tCvAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T02:53:32.284635Z"},"content_sha256":"3d2ce3ca58ff8b2b01c6cc5bf037cb3175ff28718f67ba81ce5c5ceadd016509","schema_version":"1.0","event_id":"sha256:3d2ce3ca58ff8b2b01c6cc5bf037cb3175ff28718f67ba81ce5c5ceadd016509"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:KHLIKA4AKUQDTZRUCJC2ZHENVP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TAPAS: Two-pass Approximate Adaptive Sampling for Softmax","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Li Zhang, Sally Goldman, Yu Bai","submitted_at":"2017-07-10T22:01:44Z","abstract_excerpt":"TAPAS is a novel adaptive sampling method for the softmax model. It uses a two pass sampling strategy where the examples used to approximate the gradient of the partition function are first sampled according to a squashed population distribution and then resampled adaptively using the context and current model. We describe an efficient distributed implementation of TAPAS. We show, on both synthetic data and a large real dataset, that TAPAS has low computational overhead and works well for minimizing the rank loss for multi-class classification problems with a very large label space."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1707.03073","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":""},"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":"2026-05-18T00:40:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aqwVbZUSM80Z6sWoHe9AfDd3UmJGMWyBh35Exf2DcN9rxzJxuuujia5Yxs9tzoFYqhotBImc2oq0mfmQkTNiBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T02:53:32.284973Z"},"content_sha256":"2525ee854a7dc26312bf8f59afc29d137e02b6b911e0b0f0eb65a46ca54e1093","schema_version":"1.0","event_id":"sha256:2525ee854a7dc26312bf8f59afc29d137e02b6b911e0b0f0eb65a46ca54e1093"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KHLIKA4AKUQDTZRUCJC2ZHENVP/bundle.json","state_url":"https://pith.science/pith/KHLIKA4AKUQDTZRUCJC2ZHENVP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KHLIKA4AKUQDTZRUCJC2ZHENVP/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":"2026-06-03T02:53:32Z","links":{"resolver":"https://pith.science/pith/KHLIKA4AKUQDTZRUCJC2ZHENVP","bundle":"https://pith.science/pith/KHLIKA4AKUQDTZRUCJC2ZHENVP/bundle.json","state":"https://pith.science/pith/KHLIKA4AKUQDTZRUCJC2ZHENVP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KHLIKA4AKUQDTZRUCJC2ZHENVP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:KHLIKA4AKUQDTZRUCJC2ZHENVP","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":"d56f9c466121fd17b51b4ac49934f3fa8c76eb81a2e1569b7872bd4381efbbf5","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-07-10T22:01:44Z","title_canon_sha256":"b3efe2bd834bb945fbc794b605ebb957d075d8151b6ae8a391d11a234c843934"},"schema_version":"1.0","source":{"id":"1707.03073","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1707.03073","created_at":"2026-05-18T00:40:18Z"},{"alias_kind":"arxiv_version","alias_value":"1707.03073v2","created_at":"2026-05-18T00:40:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1707.03073","created_at":"2026-05-18T00:40:18Z"},{"alias_kind":"pith_short_12","alias_value":"KHLIKA4AKUQD","created_at":"2026-05-18T12:31:24Z"},{"alias_kind":"pith_short_16","alias_value":"KHLIKA4AKUQDTZRU","created_at":"2026-05-18T12:31:24Z"},{"alias_kind":"pith_short_8","alias_value":"KHLIKA4A","created_at":"2026-05-18T12:31:24Z"}],"graph_snapshots":[{"event_id":"sha256:2525ee854a7dc26312bf8f59afc29d137e02b6b911e0b0f0eb65a46ca54e1093","target":"graph","created_at":"2026-05-18T00:40:18Z","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":"TAPAS is a novel adaptive sampling method for the softmax model. It uses a two pass sampling strategy where the examples used to approximate the gradient of the partition function are first sampled according to a squashed population distribution and then resampled adaptively using the context and current model. We describe an efficient distributed implementation of TAPAS. We show, on both synthetic data and a large real dataset, that TAPAS has low computational overhead and works well for minimizing the rank loss for multi-class classification problems with a very large label space.","authors_text":"Li Zhang, Sally Goldman, Yu Bai","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-07-10T22:01:44Z","title":"TAPAS: Two-pass Approximate Adaptive Sampling for Softmax"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1707.03073","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":"sha256:3d2ce3ca58ff8b2b01c6cc5bf037cb3175ff28718f67ba81ce5c5ceadd016509","target":"record","created_at":"2026-05-18T00:40:18Z","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":"d56f9c466121fd17b51b4ac49934f3fa8c76eb81a2e1569b7872bd4381efbbf5","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-07-10T22:01:44Z","title_canon_sha256":"b3efe2bd834bb945fbc794b605ebb957d075d8151b6ae8a391d11a234c843934"},"schema_version":"1.0","source":{"id":"1707.03073","kind":"arxiv","version":2}},"canonical_sha256":"51d6850380552039e6341245ac9c8dabc2b434a820158072cb6cbb507915d06f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"51d6850380552039e6341245ac9c8dabc2b434a820158072cb6cbb507915d06f","first_computed_at":"2026-05-18T00:40:18.885877Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:40:18.885877Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ux/3pHS+dsrodzg7baC9YyzZyQgw+NPCX69aTN6/aJkVVx/0Rxx76gIMIKYGExSEI4g6AgyKhTl6TJDe+yhCCA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:40:18.886531Z","signed_message":"canonical_sha256_bytes"},"source_id":"1707.03073","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3d2ce3ca58ff8b2b01c6cc5bf037cb3175ff28718f67ba81ce5c5ceadd016509","sha256:2525ee854a7dc26312bf8f59afc29d137e02b6b911e0b0f0eb65a46ca54e1093"],"state_sha256":"ae5df5ed5789271c8c1ca5b0385b7421b9ed9dce3c046cb340f2723b1bd779b4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1+lA2p4OkZIVKYmgx2Jh/QLlcgK2WKI9XEx6JikV57izLJYbvG1BANVqL4XjedrSWqdyjF+Qrs2rggKfBBopAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-03T02:53:32.286888Z","bundle_sha256":"36014db5a9454a9b96b806d4f547a1c339874070b4ef215bf444759a959a2de6"}}