{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W","short_pith_number":"pith:Z3QQKJ6G","schema_version":"1.0","canonical_sha256":"cee10527c6534332fce9f8c6b5dd19edb763a70e930d2a19174df6725f4d7aaf","source":{"kind":"arxiv","id":"2503.12271","version":1},"attestation_state":"computed","paper":{"title":"Reflect-DiT: Inference-Time Scaling for Text-to-Image Diffusion Transformers via In-Context Reflection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aditya Grover, Akash Gokul, Arsh Koneru, Kazuki Kozuka, Konstantinos Kallidromitis, Shufan Li, Yusuke Kato","submitted_at":"2025-03-15T21:58:12Z","abstract_excerpt":"The predominant approach to advancing text-to-image generation has been training-time scaling, where larger models are trained on more data using greater computational resources. While effective, this approach is computationally expensive, leading to growing interest in inference-time scaling to improve performance. Currently, inference-time scaling for text-to-image diffusion models is largely limited to best-of-N sampling, where multiple images are generated per prompt and a selection model chooses the best output. Inspired by the recent success of reasoning models like DeepSeek-R1 in the la"},"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":"2503.12271","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-15T21:58:12Z","cross_cats_sorted":[],"title_canon_sha256":"8fe9e0b7904785d63e348eaae23fe49e4c255c784237706165a90a7655126efb","abstract_canon_sha256":"ed3a43f1d8c71b7e0db882d5e99f0254d854547f4eb0c416d5df155450684e38"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:32:09.108021Z","signature_b64":"uLjzcvZ2mi2I06qozUlnS9VgG20Afo7AtACcbR4TmoVOEePRZpqStEYVimA8P4ELqijWsdvtt/JlVP5aBYmHDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cee10527c6534332fce9f8c6b5dd19edb763a70e930d2a19174df6725f4d7aaf","last_reissued_at":"2026-07-05T10:32:09.107399Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:32:09.107399Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reflect-DiT: Inference-Time Scaling for Text-to-Image Diffusion Transformers via In-Context Reflection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aditya Grover, Akash Gokul, Arsh Koneru, Kazuki Kozuka, Konstantinos Kallidromitis, Shufan Li, Yusuke Kato","submitted_at":"2025-03-15T21:58:12Z","abstract_excerpt":"The predominant approach to advancing text-to-image generation has been training-time scaling, where larger models are trained on more data using greater computational resources. While effective, this approach is computationally expensive, leading to growing interest in inference-time scaling to improve performance. Currently, inference-time scaling for text-to-image diffusion models is largely limited to best-of-N sampling, where multiple images are generated per prompt and a selection model chooses the best output. Inspired by the recent success of reasoning models like DeepSeek-R1 in the la"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.12271","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/2503.12271/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":"2503.12271","created_at":"2026-07-05T10:32:09.107471+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.12271v1","created_at":"2026-07-05T10:32:09.107471+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.12271","created_at":"2026-07-05T10:32:09.107471+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z3QQKJ6GKNBT","created_at":"2026-07-05T10:32:09.107471+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z3QQKJ6GKNBTF7HJ","created_at":"2026-07-05T10:32:09.107471+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z3QQKJ6G","created_at":"2026-07-05T10:32:09.107471+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.13303","citing_title":"DuET: Dual Expert Trajectories for Diffusion Image Editing","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08231","citing_title":"Test-Time Scaling in Multimodal Foundation Models: A Comprehensive Survey of Generation and Reasoning","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2512.19433","citing_title":"dMLLM-TTS: Self-Verified and Efficient Test-Time Scaling for Diffusion Multi-Modal Large Language Models","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2602.19974","citing_title":"RL-RIG: A Generative Spatial Reasoner via Intrinsic Reflection","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12724","citing_title":"Inline Critic Steers Image Editing","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W","json":"https://pith.science/pith/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W.json","graph_json":"https://pith.science/api/pith-number/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W/graph.json","events_json":"https://pith.science/api/pith-number/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W/events.json","paper":"https://pith.science/paper/Z3QQKJ6G"},"agent_actions":{"view_html":"https://pith.science/pith/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W","download_json":"https://pith.science/pith/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W.json","view_paper":"https://pith.science/paper/Z3QQKJ6G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.12271&json=true","fetch_graph":"https://pith.science/api/pith-number/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W/graph.json","fetch_events":"https://pith.science/api/pith-number/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W/action/storage_attestation","attest_author":"https://pith.science/pith/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W/action/author_attestation","sign_citation":"https://pith.science/pith/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W/action/citation_signature","submit_replication":"https://pith.science/pith/Z3QQKJ6GKNBTF7HJ7DDLLXIZ5W/action/replication_record"}},"created_at":"2026-07-05T10:32:09.107471+00:00","updated_at":"2026-07-05T10:32:09.107471+00:00"}