{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:I56QVCUPUCUGNCVJKZRPKTZ5IR","short_pith_number":"pith:I56QVCUP","schema_version":"1.0","canonical_sha256":"477d0a8a8fa0a8668aa95662f54f3d4463d40dfbd0365b4fdf6dba45fa064757","source":{"kind":"arxiv","id":"1705.05524","version":2},"attestation_state":"computed","paper":{"title":"Learning Hard Alignments with Variational Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.AI","authors_text":"Chung-Cheng Chiu, Colin Raffel, Dieterich Lawson, George Tucker, Kevin Swersky, Navdeep Jaitly","submitted_at":"2017-05-16T04:30:56Z","abstract_excerpt":"There has recently been significant interest in hard attention models for tasks such as object recognition, visual captioning and speech recognition. Hard attention can offer benefits over soft attention such as decreased computational cost, but training hard attention models can be difficult because of the discrete latent variables they introduce. Previous work used REINFORCE and Q-learning to approach these issues, but those methods can provide high-variance gradient estimates and be slow to train. In this paper, we tackle the problem of learning hard attention for a sequential task using va"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1705.05524","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2017-05-16T04:30:56Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"b6ed3f78e4d791fefc5b3f0d5f20354d9d02aec610f27a12b79f95308ec238c5","abstract_canon_sha256":"ff4b09e2c70d5b9bfdb3a86d2274d1c6c0ca7e51b9f588eb5f70f53ff8bef192"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:31:31.143165Z","signature_b64":"8aAygYUwa40f7JsI5XLPKgrtPWj/2FWjuaMRcLVmdsFFca9OyOM93S10bpyuVK1oNlBHl9L97T9hmJgmFhmUAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"477d0a8a8fa0a8668aa95662f54f3d4463d40dfbd0365b4fdf6dba45fa064757","last_reissued_at":"2026-05-18T00:31:31.142430Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:31:31.142430Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Hard Alignments with Variational Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.AI","authors_text":"Chung-Cheng Chiu, Colin Raffel, Dieterich Lawson, George Tucker, Kevin Swersky, Navdeep Jaitly","submitted_at":"2017-05-16T04:30:56Z","abstract_excerpt":"There has recently been significant interest in hard attention models for tasks such as object recognition, visual captioning and speech recognition. Hard attention can offer benefits over soft attention such as decreased computational cost, but training hard attention models can be difficult because of the discrete latent variables they introduce. Previous work used REINFORCE and Q-learning to approach these issues, but those methods can provide high-variance gradient estimates and be slow to train. In this paper, we tackle the problem of learning hard attention for a sequential task using va"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1705.05524","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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1705.05524","created_at":"2026-05-18T00:31:31.142556+00:00"},{"alias_kind":"arxiv_version","alias_value":"1705.05524v2","created_at":"2026-05-18T00:31:31.142556+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1705.05524","created_at":"2026-05-18T00:31:31.142556+00:00"},{"alias_kind":"pith_short_12","alias_value":"I56QVCUPUCUG","created_at":"2026-05-18T12:31:21.493067+00:00"},{"alias_kind":"pith_short_16","alias_value":"I56QVCUPUCUGNCVJ","created_at":"2026-05-18T12:31:21.493067+00:00"},{"alias_kind":"pith_short_8","alias_value":"I56QVCUP","created_at":"2026-05-18T12:31:21.493067+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/I56QVCUPUCUGNCVJKZRPKTZ5IR","json":"https://pith.science/pith/I56QVCUPUCUGNCVJKZRPKTZ5IR.json","graph_json":"https://pith.science/api/pith-number/I56QVCUPUCUGNCVJKZRPKTZ5IR/graph.json","events_json":"https://pith.science/api/pith-number/I56QVCUPUCUGNCVJKZRPKTZ5IR/events.json","paper":"https://pith.science/paper/I56QVCUP"},"agent_actions":{"view_html":"https://pith.science/pith/I56QVCUPUCUGNCVJKZRPKTZ5IR","download_json":"https://pith.science/pith/I56QVCUPUCUGNCVJKZRPKTZ5IR.json","view_paper":"https://pith.science/paper/I56QVCUP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1705.05524&json=true","fetch_graph":"https://pith.science/api/pith-number/I56QVCUPUCUGNCVJKZRPKTZ5IR/graph.json","fetch_events":"https://pith.science/api/pith-number/I56QVCUPUCUGNCVJKZRPKTZ5IR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I56QVCUPUCUGNCVJKZRPKTZ5IR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I56QVCUPUCUGNCVJKZRPKTZ5IR/action/storage_attestation","attest_author":"https://pith.science/pith/I56QVCUPUCUGNCVJKZRPKTZ5IR/action/author_attestation","sign_citation":"https://pith.science/pith/I56QVCUPUCUGNCVJKZRPKTZ5IR/action/citation_signature","submit_replication":"https://pith.science/pith/I56QVCUPUCUGNCVJKZRPKTZ5IR/action/replication_record"}},"created_at":"2026-05-18T00:31:31.142556+00:00","updated_at":"2026-05-18T00:31:31.142556+00:00"}