{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:EM3FJITNE6ISPGXOZDNQ3XXNJG","short_pith_number":"pith:EM3FJITN","schema_version":"1.0","canonical_sha256":"233654a26d2791279aeec8db0ddeed49ba362c217843ba32fb756556f27fc44f","source":{"kind":"arxiv","id":"2102.12092","version":2},"attestation_state":"computed","paper":{"title":"Zero-Shot Text-to-Image Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A transformer that models text and image tokens as one autoregressive stream achieves competitive zero-shot text-to-image generation at sufficient scale.","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Aditya Ramesh, Alec Radford, Chelsea Voss, Gabriel Goh, Ilya Sutskever, Mark Chen, Mikhail Pavlov, Scott Gray","submitted_at":"2021-02-24T06:42:31Z","abstract_excerpt":"Text-to-image generation has traditionally focused on finding better modeling assumptions for training on a fixed dataset. These assumptions might involve complex architectures, auxiliary losses, or side information such as object part labels or segmentation masks supplied during training. We describe a simple approach for this task based on a transformer that autoregressively models the text and image tokens as a single stream of data. With sufficient data and scale, our approach is competitive with previous domain-specific models when evaluated in a zero-shot fashion."},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2102.12092","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-02-24T06:42:31Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d8fed5258329bcb50974f59ef498b8b49dee66857162c9241f9695116f4ebbec","abstract_canon_sha256":"7b1b5ca549ba2968b6bbe3695aa883be7800890084b682e1f3e752983ebc93ab"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T04:19:54.695463Z","signature_b64":"eHsVQDwY2XUVi9674KmGzDEz78GtLX14VYyCPLV6tvr3pFWrBx0mCeJOKQW+d42UKP4gKEKuLI9eOpivFBF2Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"233654a26d2791279aeec8db0ddeed49ba362c217843ba32fb756556f27fc44f","last_reissued_at":"2026-05-18T04:19:54.694687Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T04:19:54.694687Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Zero-Shot Text-to-Image Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A transformer that models text and image tokens as one autoregressive stream achieves competitive zero-shot text-to-image generation at sufficient scale.","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Aditya Ramesh, Alec Radford, Chelsea Voss, Gabriel Goh, Ilya Sutskever, Mark Chen, Mikhail Pavlov, Scott Gray","submitted_at":"2021-02-24T06:42:31Z","abstract_excerpt":"Text-to-image generation has traditionally focused on finding better modeling assumptions for training on a fixed dataset. These assumptions might involve complex architectures, auxiliary losses, or side information such as object part labels or segmentation masks supplied during training. We describe a simple approach for this task based on a transformer that autoregressively models the text and image tokens as a single stream of data. With sufficient data and scale, our approach is competitive with previous domain-specific models when evaluated in a zero-shot fashion."},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"With sufficient data and scale, our approach is competitive with previous domain-specific models when evaluated in a zero-shot fashion.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That simply increasing model size and data volume will continue to close the gap to specialized models without introducing new failure modes or requiring additional inductive biases.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A transformer autoregressively models text and image tokens as one stream and produces competitive zero-shot text-to-image results at sufficient scale.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A transformer that models text and image tokens as one autoregressive stream achieves competitive zero-shot text-to-image generation at sufficient scale.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"13db172eac172da28dfbbecbfd58e7c97ef70b1dabb61fe7a9acc90130fd22bc"},"source":{"id":"2102.12092","kind":"arxiv","version":2},"verdict":{"id":"d06740f6-2e1e-4b26-9524-ef4eedfd647c","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T22:22:00.876735Z","strongest_claim":"With sufficient data and scale, our approach is competitive with previous domain-specific models when evaluated in a zero-shot fashion.","one_line_summary":"A transformer autoregressively models text and image tokens as one stream and produces competitive zero-shot text-to-image results at sufficient scale.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That simply increasing model size and data volume will continue to close the gap to specialized models without introducing new failure modes or requiring additional inductive biases.","pith_extraction_headline":"A transformer that models text and image tokens as one autoregressive stream achieves competitive zero-shot text-to-image generation at sufficient scale."},"references":{"count":23,"sample":[{"doi":"","year":2015,"title":"Bowman et al","work_id":"52301570-311d-400c-a35e-e91c2dafc161","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Using a linear annealing schedule for this typically led to divergence","work_id":"a760669a-0fff-48d1-8242-87dac551d654","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"row, column, row, row","work_id":"0336e6b0-451d-47dc-8d97-1a46d4107b72","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Our model uses 128 gradient scales, one for each of its resblocks","work_id":"acbee4e8-41fd-48e1-b387-4b6685b9e6e2","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"In particular, store all gains, biases, embeddings, and unembeddings in 32-bit precision, with 32-bit gradients (including for remote communication) and 32-bit Adam moments","work_id":"2cbd3bc9-f9b0-4459-902d-4ad616b9d452","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":23,"snapshot_sha256":"20e3e9821b8d8cdb6d7d5423026327248332c552ac100fa4abd9e4fd6b4d7666","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"d12df4dfee20ea1cf31da99d9d8fe14ff7462fb2fce8615d7054987eb1475f96"},"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":"2102.12092","created_at":"2026-05-18T04:19:54.694831+00:00"},{"alias_kind":"arxiv_version","alias_value":"2102.12092v2","created_at":"2026-05-18T04:19:54.694831+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.12092","created_at":"2026-05-18T04:19:54.694831+00:00"},{"alias_kind":"pith_short_12","alias_value":"EM3FJITNE6IS","created_at":"2026-05-18T12:33:33.725879+00:00"},{"alias_kind":"pith_short_16","alias_value":"EM3FJITNE6ISPGXO","created_at":"2026-05-18T12:33:33.725879+00:00"},{"alias_kind":"pith_short_8","alias_value":"EM3FJITN","created_at":"2026-05-18T12:33:33.725879+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":37,"internal_anchor_count":37,"sample":[{"citing_arxiv_id":"2606.24021","citing_title":"Token-to-Token Alignment of Text Embeddings for Semantic Blending","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2606.08841","citing_title":"ZIPP:Zero-shot Image Personalization from Personas","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2606.02305","citing_title":"Mapping Whisper Representations to Human ECoG Responses with Interpretable Time-Resolved Neural Encoding","ref_index":126,"is_internal_anchor":true},{"citing_arxiv_id":"2208.14649","citing_title":"DetailCLIP: Injecting Image Details into CLIP's Feature Space","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20866","citing_title":"LOSCAR-SGD: Local SGD with Communication-Computation Overlap and Delay-Corrected Sparse Model Averaging","ref_index":143,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18174","citing_title":"Ringmaster LMO: Asynchronous Linear Minimization Oracle Momentum Method","ref_index":141,"is_internal_anchor":true},{"citing_arxiv_id":"2506.20911","citing_title":"FaSTA$^*$: Fast-Slow Toolpath Agent with Subroutine Mining for Efficient Multi-turn Image Editing","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2106.01345","citing_title":"Decision Transformer: Reinforcement Learning via Sequence Modeling","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2105.04663","citing_title":"GSPMD: General and Scalable Parallelization for ML Computation Graphs","ref_index":31,"is_internal_anchor":true},{"citing_arxiv_id":"2411.04996","citing_title":"Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models","ref_index":31,"is_internal_anchor":true},{"citing_arxiv_id":"2305.02463","citing_title":"Shap-E: Generating Conditional 3D Implicit Functions","ref_index":49,"is_internal_anchor":true},{"citing_arxiv_id":"2111.11432","citing_title":"Florence: A New Foundation Model for Computer Vision","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2309.16671","citing_title":"Demystifying CLIP Data","ref_index":104,"is_internal_anchor":true},{"citing_arxiv_id":"2201.10005","citing_title":"Text and Code Embeddings by Contrastive Pre-Training","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2312.14125","citing_title":"VideoPoet: A Large Language Model for Zero-Shot Video Generation","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2604.02482","citing_title":"SEDGE: Structural Extrapolated Data Generation","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2604.02482","citing_title":"SEDGE: Structural Extrapolated Data Generation","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2104.10157","citing_title":"VideoGPT: Video Generation using VQ-VAE and Transformers","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2210.08402","citing_title":"LAION-5B: An open large-scale dataset for training next generation image-text models","ref_index":61,"is_internal_anchor":true},{"citing_arxiv_id":"2208.07339","citing_title":"LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale","ref_index":66,"is_internal_anchor":true},{"citing_arxiv_id":"2106.08254","citing_title":"BEiT: BERT Pre-Training of Image Transformers","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2105.05233","citing_title":"Diffusion Models Beat GANs on Image Synthesis","ref_index":50,"is_internal_anchor":true},{"citing_arxiv_id":"2303.15389","citing_title":"EVA-CLIP: Improved Training Techniques for CLIP at Scale","ref_index":42,"is_internal_anchor":true},{"citing_arxiv_id":"2111.02114","citing_title":"LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08871","citing_title":"Rennala MVR: Improved Time Complexity for Parallel Stochastic Optimization via Momentum-Based Variance Reduction","ref_index":139,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EM3FJITNE6ISPGXOZDNQ3XXNJG","json":"https://pith.science/pith/EM3FJITNE6ISPGXOZDNQ3XXNJG.json","graph_json":"https://pith.science/api/pith-number/EM3FJITNE6ISPGXOZDNQ3XXNJG/graph.json","events_json":"https://pith.science/api/pith-number/EM3FJITNE6ISPGXOZDNQ3XXNJG/events.json","paper":"https://pith.science/paper/EM3FJITN"},"agent_actions":{"view_html":"https://pith.science/pith/EM3FJITNE6ISPGXOZDNQ3XXNJG","download_json":"https://pith.science/pith/EM3FJITNE6ISPGXOZDNQ3XXNJG.json","view_paper":"https://pith.science/paper/EM3FJITN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2102.12092&json=true","fetch_graph":"https://pith.science/api/pith-number/EM3FJITNE6ISPGXOZDNQ3XXNJG/graph.json","fetch_events":"https://pith.science/api/pith-number/EM3FJITNE6ISPGXOZDNQ3XXNJG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EM3FJITNE6ISPGXOZDNQ3XXNJG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EM3FJITNE6ISPGXOZDNQ3XXNJG/action/storage_attestation","attest_author":"https://pith.science/pith/EM3FJITNE6ISPGXOZDNQ3XXNJG/action/author_attestation","sign_citation":"https://pith.science/pith/EM3FJITNE6ISPGXOZDNQ3XXNJG/action/citation_signature","submit_replication":"https://pith.science/pith/EM3FJITNE6ISPGXOZDNQ3XXNJG/action/replication_record"}},"created_at":"2026-05-18T04:19:54.694831+00:00","updated_at":"2026-05-18T04:19:54.694831+00:00"}