{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:Z7Y4YH7FVFQORYWDXLMORB3AWD","short_pith_number":"pith:Z7Y4YH7F","schema_version":"1.0","canonical_sha256":"cff1cc1fe5a960e8e2c3bad8e88760b0c63bb3acc98388980bbbc8f4dd3cd238","source":{"kind":"arxiv","id":"2305.18619","version":1},"attestation_state":"computed","paper":{"title":"Likelihood-Based Diffusion Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Ishaan Gulrajani, Tatsunori B. Hashimoto","submitted_at":"2023-05-30T16:43:31Z","abstract_excerpt":"Despite a growing interest in diffusion-based language models, existing work has not shown that these models can attain nontrivial likelihoods on standard language modeling benchmarks. In this work, we take the first steps towards closing the likelihood gap between autoregressive and diffusion-based language models, with the goal of building and releasing a diffusion model which outperforms a small but widely-known autoregressive model. We pursue this goal through algorithmic improvements, scaling laws, and increased compute. On the algorithmic front, we introduce several methodological improv"},"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":"2305.18619","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-30T16:43:31Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"89abfded2dd1b5a9c061348f1293fd9e53abfd5461a864e423b0208d0f86a9b0","abstract_canon_sha256":"ef19f9692892c7bcdcb5d94490c23169404589392ad56eb36eab7fc72f2cc041"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:15:02.964844Z","signature_b64":"LUgRmv3xMT5X7qFWD+6kzVVn+XxjZ+6rzyqj8kfxQHXNzCJwlXzTXEHJquG2o5LGfF5NRErnxkjtnJiiuguODg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cff1cc1fe5a960e8e2c3bad8e88760b0c63bb3acc98388980bbbc8f4dd3cd238","last_reissued_at":"2026-07-05T06:15:02.964506Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:15:02.964506Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Likelihood-Based Diffusion Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Ishaan Gulrajani, Tatsunori B. Hashimoto","submitted_at":"2023-05-30T16:43:31Z","abstract_excerpt":"Despite a growing interest in diffusion-based language models, existing work has not shown that these models can attain nontrivial likelihoods on standard language modeling benchmarks. In this work, we take the first steps towards closing the likelihood gap between autoregressive and diffusion-based language models, with the goal of building and releasing a diffusion model which outperforms a small but widely-known autoregressive model. We pursue this goal through algorithmic improvements, scaling laws, and increased compute. On the algorithmic front, we introduce several methodological improv"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.18619","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/2305.18619/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":"2305.18619","created_at":"2026-07-05T06:15:02.964562+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.18619v1","created_at":"2026-07-05T06:15:02.964562+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.18619","created_at":"2026-07-05T06:15:02.964562+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z7Y4YH7FVFQO","created_at":"2026-07-05T06:15:02.964562+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z7Y4YH7FVFQORYWD","created_at":"2026-07-05T06:15:02.964562+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z7Y4YH7F","created_at":"2026-07-05T06:15:02.964562+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/Z7Y4YH7FVFQORYWDXLMORB3AWD","json":"https://pith.science/pith/Z7Y4YH7FVFQORYWDXLMORB3AWD.json","graph_json":"https://pith.science/api/pith-number/Z7Y4YH7FVFQORYWDXLMORB3AWD/graph.json","events_json":"https://pith.science/api/pith-number/Z7Y4YH7FVFQORYWDXLMORB3AWD/events.json","paper":"https://pith.science/paper/Z7Y4YH7F"},"agent_actions":{"view_html":"https://pith.science/pith/Z7Y4YH7FVFQORYWDXLMORB3AWD","download_json":"https://pith.science/pith/Z7Y4YH7FVFQORYWDXLMORB3AWD.json","view_paper":"https://pith.science/paper/Z7Y4YH7F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.18619&json=true","fetch_graph":"https://pith.science/api/pith-number/Z7Y4YH7FVFQORYWDXLMORB3AWD/graph.json","fetch_events":"https://pith.science/api/pith-number/Z7Y4YH7FVFQORYWDXLMORB3AWD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z7Y4YH7FVFQORYWDXLMORB3AWD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z7Y4YH7FVFQORYWDXLMORB3AWD/action/storage_attestation","attest_author":"https://pith.science/pith/Z7Y4YH7FVFQORYWDXLMORB3AWD/action/author_attestation","sign_citation":"https://pith.science/pith/Z7Y4YH7FVFQORYWDXLMORB3AWD/action/citation_signature","submit_replication":"https://pith.science/pith/Z7Y4YH7FVFQORYWDXLMORB3AWD/action/replication_record"}},"created_at":"2026-07-05T06:15:02.964562+00:00","updated_at":"2026-07-05T06:15:02.964562+00:00"}