{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:NFECBVGPDSTGCLHGTEV5O6QDNK","short_pith_number":"pith:NFECBVGP","schema_version":"1.0","canonical_sha256":"694820d4cf1ca6612ce6992bd77a036a8e0ea142477d813a224da2b032292240","source":{"kind":"arxiv","id":"2305.16944","version":3},"attestation_state":"computed","paper":{"title":"Learning to Imagine: Visually-Augmented Natural Language Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ji-Rong Wen, Junyi Li, Tianyi Tang, Wayne Xin Zhao, Yifan Du, Yushuo Chen","submitted_at":"2023-05-26T13:59:45Z","abstract_excerpt":"People often imagine relevant scenes to aid in the writing process. In this work, we aim to utilize visual information for composition in the same manner as humans. We propose a method, LIVE, that makes pre-trained language models (PLMs) Learn to Imagine for Visuallyaugmented natural language gEneration. First, we imagine the scene based on the text: we use a diffusion model to synthesize high-quality images conditioned on the input texts. Second, we use CLIP to determine whether the text can evoke the imagination in a posterior way. Finally, our imagination is dynamic, and we conduct synthesi"},"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":"2305.16944","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-26T13:59:45Z","cross_cats_sorted":[],"title_canon_sha256":"e6b0b09327b599fa1c37207e8d66cba39d8d10830598d9689a57ed0d34f9986e","abstract_canon_sha256":"d4c2d1aff44d638aa77d196fcb8134ec566d102cf32b9e495a4ae2b34a9c8b72"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:20:44.400092Z","signature_b64":"yvPZPuSwrB1ixDbadUZB6IDh03dYYj7aaSRbwvkacL9tTNJSDm4zpTSEXFZrStF8faD3Mm4tI78uBBAqUTTODg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"694820d4cf1ca6612ce6992bd77a036a8e0ea142477d813a224da2b032292240","last_reissued_at":"2026-07-05T06:20:44.399614Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:20:44.399614Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Imagine: Visually-Augmented Natural Language Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ji-Rong Wen, Junyi Li, Tianyi Tang, Wayne Xin Zhao, Yifan Du, Yushuo Chen","submitted_at":"2023-05-26T13:59:45Z","abstract_excerpt":"People often imagine relevant scenes to aid in the writing process. In this work, we aim to utilize visual information for composition in the same manner as humans. We propose a method, LIVE, that makes pre-trained language models (PLMs) Learn to Imagine for Visuallyaugmented natural language gEneration. First, we imagine the scene based on the text: we use a diffusion model to synthesize high-quality images conditioned on the input texts. Second, we use CLIP to determine whether the text can evoke the imagination in a posterior way. Finally, our imagination is dynamic, and we conduct synthesi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.16944","kind":"arxiv","version":3},"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/2305.16944/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":"2305.16944","created_at":"2026-07-05T06:20:44.399679+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.16944v3","created_at":"2026-07-05T06:20:44.399679+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.16944","created_at":"2026-07-05T06:20:44.399679+00:00"},{"alias_kind":"pith_short_12","alias_value":"NFECBVGPDSTG","created_at":"2026-07-05T06:20:44.399679+00:00"},{"alias_kind":"pith_short_16","alias_value":"NFECBVGPDSTGCLHG","created_at":"2026-07-05T06:20:44.399679+00:00"},{"alias_kind":"pith_short_8","alias_value":"NFECBVGP","created_at":"2026-07-05T06:20:44.399679+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.00684","citing_title":"Paint Outside the Box: Synthesizing and Selecting Training Data for Visual Grounding","ref_index":72,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NFECBVGPDSTGCLHGTEV5O6QDNK","json":"https://pith.science/pith/NFECBVGPDSTGCLHGTEV5O6QDNK.json","graph_json":"https://pith.science/api/pith-number/NFECBVGPDSTGCLHGTEV5O6QDNK/graph.json","events_json":"https://pith.science/api/pith-number/NFECBVGPDSTGCLHGTEV5O6QDNK/events.json","paper":"https://pith.science/paper/NFECBVGP"},"agent_actions":{"view_html":"https://pith.science/pith/NFECBVGPDSTGCLHGTEV5O6QDNK","download_json":"https://pith.science/pith/NFECBVGPDSTGCLHGTEV5O6QDNK.json","view_paper":"https://pith.science/paper/NFECBVGP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.16944&json=true","fetch_graph":"https://pith.science/api/pith-number/NFECBVGPDSTGCLHGTEV5O6QDNK/graph.json","fetch_events":"https://pith.science/api/pith-number/NFECBVGPDSTGCLHGTEV5O6QDNK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NFECBVGPDSTGCLHGTEV5O6QDNK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NFECBVGPDSTGCLHGTEV5O6QDNK/action/storage_attestation","attest_author":"https://pith.science/pith/NFECBVGPDSTGCLHGTEV5O6QDNK/action/author_attestation","sign_citation":"https://pith.science/pith/NFECBVGPDSTGCLHGTEV5O6QDNK/action/citation_signature","submit_replication":"https://pith.science/pith/NFECBVGPDSTGCLHGTEV5O6QDNK/action/replication_record"}},"created_at":"2026-07-05T06:20:44.399679+00:00","updated_at":"2026-07-05T06:20:44.399679+00:00"}