{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:TEQTNW2NMVALX6B6TDC6XJAVPB","short_pith_number":"pith:TEQTNW2N","schema_version":"1.0","canonical_sha256":"992136db4d6540bbf83e98c5eba415784f9eadb254bb7b26dd04e1c2484e5d87","source":{"kind":"arxiv","id":"2305.14771","version":2},"attestation_state":"computed","paper":{"title":"David helps Goliath: Inference-Time Collaboration Between Small Specialized and Large General Diffusion LMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Marjan Ghazvininejad, Sachin Kumar, Xiaochuang Han, Yulia Tsvetkov","submitted_at":"2023-05-24T06:22:14Z","abstract_excerpt":"Diffusion-based language models are emerging as a promising alternative to autoregressive LMs: they approach the competence of autoregressive LMs while offering nuanced controllability at inference time. While autoregressive LMs have benefited immensely from scaling and instruction-based learning, existing studies of diffusion LMs have been conducted on a smaller scale. Starting with a recently proposed diffusion model SSD-LM, in this work we first explore methods to scale it from 0.4B to 13B parameters, proposing techniques to improve its training and inference efficiency, and to finetune the"},"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":"2305.14771","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-24T06:22:14Z","cross_cats_sorted":[],"title_canon_sha256":"a56c26a355d11a500899e027951481864941dad05c774a81eb869a0df1936efb","abstract_canon_sha256":"91f562fae9ff23c3e53b26dcc5f503049ba25e03356448c817bc564db94cff4e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:44:54.255193Z","signature_b64":"Es7tMZhPAJJu8xX3Wy49kPPJFI3YFYAwawVykJt4z2cmmtl07R0VLEKDUbuQi2R5Q7MalHcpf9OjZ3vW/NDLBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"992136db4d6540bbf83e98c5eba415784f9eadb254bb7b26dd04e1c2484e5d87","last_reissued_at":"2026-07-05T07:44:54.254696Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:44:54.254696Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"David helps Goliath: Inference-Time Collaboration Between Small Specialized and Large General Diffusion LMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Marjan Ghazvininejad, Sachin Kumar, Xiaochuang Han, Yulia Tsvetkov","submitted_at":"2023-05-24T06:22:14Z","abstract_excerpt":"Diffusion-based language models are emerging as a promising alternative to autoregressive LMs: they approach the competence of autoregressive LMs while offering nuanced controllability at inference time. While autoregressive LMs have benefited immensely from scaling and instruction-based learning, existing studies of diffusion LMs have been conducted on a smaller scale. Starting with a recently proposed diffusion model SSD-LM, in this work we first explore methods to scale it from 0.4B to 13B parameters, proposing techniques to improve its training and inference efficiency, and to finetune the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14771","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2305.14771/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":"2305.14771","created_at":"2026-07-05T07:44:54.254751+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.14771v2","created_at":"2026-07-05T07:44:54.254751+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14771","created_at":"2026-07-05T07:44:54.254751+00:00"},{"alias_kind":"pith_short_12","alias_value":"TEQTNW2NMVAL","created_at":"2026-07-05T07:44:54.254751+00:00"},{"alias_kind":"pith_short_16","alias_value":"TEQTNW2NMVALX6B6","created_at":"2026-07-05T07:44:54.254751+00:00"},{"alias_kind":"pith_short_8","alias_value":"TEQTNW2N","created_at":"2026-07-05T07:44:54.254751+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"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":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TEQTNW2NMVALX6B6TDC6XJAVPB","json":"https://pith.science/pith/TEQTNW2NMVALX6B6TDC6XJAVPB.json","graph_json":"https://pith.science/api/pith-number/TEQTNW2NMVALX6B6TDC6XJAVPB/graph.json","events_json":"https://pith.science/api/pith-number/TEQTNW2NMVALX6B6TDC6XJAVPB/events.json","paper":"https://pith.science/paper/TEQTNW2N"},"agent_actions":{"view_html":"https://pith.science/pith/TEQTNW2NMVALX6B6TDC6XJAVPB","download_json":"https://pith.science/pith/TEQTNW2NMVALX6B6TDC6XJAVPB.json","view_paper":"https://pith.science/paper/TEQTNW2N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.14771&json=true","fetch_graph":"https://pith.science/api/pith-number/TEQTNW2NMVALX6B6TDC6XJAVPB/graph.json","fetch_events":"https://pith.science/api/pith-number/TEQTNW2NMVALX6B6TDC6XJAVPB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TEQTNW2NMVALX6B6TDC6XJAVPB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TEQTNW2NMVALX6B6TDC6XJAVPB/action/storage_attestation","attest_author":"https://pith.science/pith/TEQTNW2NMVALX6B6TDC6XJAVPB/action/author_attestation","sign_citation":"https://pith.science/pith/TEQTNW2NMVALX6B6TDC6XJAVPB/action/citation_signature","submit_replication":"https://pith.science/pith/TEQTNW2NMVALX6B6TDC6XJAVPB/action/replication_record"}},"created_at":"2026-07-05T07:44:54.254751+00:00","updated_at":"2026-07-05T07:44:54.254751+00:00"}