{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ODNYPODDBRCBCTATMIBJFZPVAH","short_pith_number":"pith:ODNYPODD","schema_version":"1.0","canonical_sha256":"70db87b8630c44114c13620292e5f501d6128b7b10ad375826b80f2b14076976","source":{"kind":"arxiv","id":"2410.21357","version":4},"attestation_state":"computed","paper":{"title":"Energy-Based Diffusion Language Models for Text Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Arash Vahdat, Jure Leskovec, Karsten Kreis, Minkai Xu, Stefano Ermon, Tomas Geffner, Weili Nie, Yilun Xu","submitted_at":"2024-10-28T17:25:56Z","abstract_excerpt":"Despite remarkable progress in autoregressive language models, alternative generative paradigms beyond left-to-right generation are still being actively explored. Discrete diffusion models, with the capacity for parallel generation, have recently emerged as a promising alternative. Unfortunately, these models still underperform the autoregressive counterparts, with the performance gap increasing when reducing the number of sampling steps. Our analysis reveals that this degradation is a consequence of an imperfect approximation used by diffusion models. In this work, we propose Energy-based Dif"},"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":"2410.21357","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-28T17:25:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9d1921e6dc697eac540106ab6076bb5e315ba3b9881546e8e9b2cfe5f8a2653d","abstract_canon_sha256":"42c37ff91d036321c9ba09f32a9aa7a4b046969b294783952990c77ab3ab074b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:26:09.624789Z","signature_b64":"bDL3D6BDJreBNSjN/P2V3UQsjlLRPwIy7kvmXVSWvHC97durbEjdSPDrysqQdEgexm8hjPjyT7q5L53VNphjDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"70db87b8630c44114c13620292e5f501d6128b7b10ad375826b80f2b14076976","last_reissued_at":"2026-07-05T10:26:09.623939Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:26:09.623939Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Energy-Based Diffusion Language Models for Text Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Arash Vahdat, Jure Leskovec, Karsten Kreis, Minkai Xu, Stefano Ermon, Tomas Geffner, Weili Nie, Yilun Xu","submitted_at":"2024-10-28T17:25:56Z","abstract_excerpt":"Despite remarkable progress in autoregressive language models, alternative generative paradigms beyond left-to-right generation are still being actively explored. Discrete diffusion models, with the capacity for parallel generation, have recently emerged as a promising alternative. Unfortunately, these models still underperform the autoregressive counterparts, with the performance gap increasing when reducing the number of sampling steps. Our analysis reveals that this degradation is a consequence of an imperfect approximation used by diffusion models. In this work, we propose Energy-based Dif"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.21357","kind":"arxiv","version":4},"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/2410.21357/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":"2410.21357","created_at":"2026-07-05T10:26:09.624050+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.21357v4","created_at":"2026-07-05T10:26:09.624050+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.21357","created_at":"2026-07-05T10:26:09.624050+00:00"},{"alias_kind":"pith_short_12","alias_value":"ODNYPODDBRCB","created_at":"2026-07-05T10:26:09.624050+00:00"},{"alias_kind":"pith_short_16","alias_value":"ODNYPODDBRCBCTAT","created_at":"2026-07-05T10:26:09.624050+00:00"},{"alias_kind":"pith_short_8","alias_value":"ODNYPODD","created_at":"2026-07-05T10:26:09.624050+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01775","citing_title":"Set Diffusion: Interpolating Token Orderings Between Autoregression and Diffusion for Fast and Flexible Decoding","ref_index":113,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09159","citing_title":"Unified Energy for Invariant and Independent Decoding in Diffusion Language Models","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07079","citing_title":"AsyncPatch Diffusion: spatially-flexible image generation","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2602.16169","citing_title":"Discrete Stochastic Localization for Non-autoregressive Generation","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2603.13419","citing_title":"Diffusion Models Memorize in Training -- and Generalize in Inference","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2505.17384","citing_title":"Variational Autoencoding Discrete Diffusion with Enhanced Dimensional Correlations Modeling","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2505.22618","citing_title":"Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06548","citing_title":"Continuous Latent Diffusion Language Model","ref_index":99,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06778","citing_title":"RichMap: A Reachability Map Balancing Precision, Efficiency, and Flexibility for Rich Robot Manipulation Tasks","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ODNYPODDBRCBCTATMIBJFZPVAH","json":"https://pith.science/pith/ODNYPODDBRCBCTATMIBJFZPVAH.json","graph_json":"https://pith.science/api/pith-number/ODNYPODDBRCBCTATMIBJFZPVAH/graph.json","events_json":"https://pith.science/api/pith-number/ODNYPODDBRCBCTATMIBJFZPVAH/events.json","paper":"https://pith.science/paper/ODNYPODD"},"agent_actions":{"view_html":"https://pith.science/pith/ODNYPODDBRCBCTATMIBJFZPVAH","download_json":"https://pith.science/pith/ODNYPODDBRCBCTATMIBJFZPVAH.json","view_paper":"https://pith.science/paper/ODNYPODD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.21357&json=true","fetch_graph":"https://pith.science/api/pith-number/ODNYPODDBRCBCTATMIBJFZPVAH/graph.json","fetch_events":"https://pith.science/api/pith-number/ODNYPODDBRCBCTATMIBJFZPVAH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ODNYPODDBRCBCTATMIBJFZPVAH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ODNYPODDBRCBCTATMIBJFZPVAH/action/storage_attestation","attest_author":"https://pith.science/pith/ODNYPODDBRCBCTATMIBJFZPVAH/action/author_attestation","sign_citation":"https://pith.science/pith/ODNYPODDBRCBCTATMIBJFZPVAH/action/citation_signature","submit_replication":"https://pith.science/pith/ODNYPODDBRCBCTATMIBJFZPVAH/action/replication_record"}},"created_at":"2026-07-05T10:26:09.624050+00:00","updated_at":"2026-07-05T10:26:09.624050+00:00"}