{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:A5HKX26NWKQQAZPPQJI77TKPEQ","short_pith_number":"pith:A5HKX26N","schema_version":"1.0","canonical_sha256":"074eabebcdb2a10065ef8251ffcd4f2403741c0ba8d703906fca5efa8b53f2e6","source":{"kind":"arxiv","id":"2206.12840","version":1},"attestation_state":"computed","paper":{"title":"Your Autoregressive Generative Model Can be Better If You Treat It as an Energy-Based One","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Bo Li, Dongsheng Li, Hengzhi Pei, Kaitao Song, Tong Che, Yezhen Wang, Yoshua Bengio","submitted_at":"2022-06-26T10:58:41Z","abstract_excerpt":"Autoregressive generative models are commonly used, especially for those tasks involving sequential data. They have, however, been plagued by a slew of inherent flaws due to the intrinsic characteristics of chain-style conditional modeling (e.g., exposure bias or lack of long-range coherence), severely limiting their ability to model distributions properly. In this paper, we propose a unique method termed E-ARM for training autoregressive generative models that takes advantage of a well-designed energy-based learning objective. By leveraging the extra degree of freedom of the softmax operation"},"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":"2206.12840","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-06-26T10:58:41Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"1c2dc94bf0f911fef4f00df495629c8cd8f307b183e051f23f5272d8452e96ea","abstract_canon_sha256":"3428a730b38ec5e9445d2b71459b08af08178aaa61c6847b22a8ff71fc97ad55"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:34:59.828858Z","signature_b64":"zjzVaoReXx1+m8/dSCtlQjOhsPGK4tbToAZV7y01fMPKuGSwm0YLPz1Wv5xor5BvOBpC2TclR76ea/7zJzdqBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"074eabebcdb2a10065ef8251ffcd4f2403741c0ba8d703906fca5efa8b53f2e6","last_reissued_at":"2026-07-05T04:34:59.828380Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:34:59.828380Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Your Autoregressive Generative Model Can be Better If You Treat It as an Energy-Based One","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Bo Li, Dongsheng Li, Hengzhi Pei, Kaitao Song, Tong Che, Yezhen Wang, Yoshua Bengio","submitted_at":"2022-06-26T10:58:41Z","abstract_excerpt":"Autoregressive generative models are commonly used, especially for those tasks involving sequential data. They have, however, been plagued by a slew of inherent flaws due to the intrinsic characteristics of chain-style conditional modeling (e.g., exposure bias or lack of long-range coherence), severely limiting their ability to model distributions properly. In this paper, we propose a unique method termed E-ARM for training autoregressive generative models that takes advantage of a well-designed energy-based learning objective. By leveraging the extra degree of freedom of the softmax operation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.12840","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/2206.12840/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2206.12840","created_at":"2026-07-05T04:34:59.828440+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.12840v1","created_at":"2026-07-05T04:34:59.828440+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.12840","created_at":"2026-07-05T04:34:59.828440+00:00"},{"alias_kind":"pith_short_12","alias_value":"A5HKX26NWKQQ","created_at":"2026-07-05T04:34:59.828440+00:00"},{"alias_kind":"pith_short_16","alias_value":"A5HKX26NWKQQAZPP","created_at":"2026-07-05T04:34:59.828440+00:00"},{"alias_kind":"pith_short_8","alias_value":"A5HKX26N","created_at":"2026-07-05T04:34:59.828440+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/A5HKX26NWKQQAZPPQJI77TKPEQ","json":"https://pith.science/pith/A5HKX26NWKQQAZPPQJI77TKPEQ.json","graph_json":"https://pith.science/api/pith-number/A5HKX26NWKQQAZPPQJI77TKPEQ/graph.json","events_json":"https://pith.science/api/pith-number/A5HKX26NWKQQAZPPQJI77TKPEQ/events.json","paper":"https://pith.science/paper/A5HKX26N"},"agent_actions":{"view_html":"https://pith.science/pith/A5HKX26NWKQQAZPPQJI77TKPEQ","download_json":"https://pith.science/pith/A5HKX26NWKQQAZPPQJI77TKPEQ.json","view_paper":"https://pith.science/paper/A5HKX26N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.12840&json=true","fetch_graph":"https://pith.science/api/pith-number/A5HKX26NWKQQAZPPQJI77TKPEQ/graph.json","fetch_events":"https://pith.science/api/pith-number/A5HKX26NWKQQAZPPQJI77TKPEQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A5HKX26NWKQQAZPPQJI77TKPEQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A5HKX26NWKQQAZPPQJI77TKPEQ/action/storage_attestation","attest_author":"https://pith.science/pith/A5HKX26NWKQQAZPPQJI77TKPEQ/action/author_attestation","sign_citation":"https://pith.science/pith/A5HKX26NWKQQAZPPQJI77TKPEQ/action/citation_signature","submit_replication":"https://pith.science/pith/A5HKX26NWKQQAZPPQJI77TKPEQ/action/replication_record"}},"created_at":"2026-07-05T04:34:59.828440+00:00","updated_at":"2026-07-05T04:34:59.828440+00:00"}