{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LJ5DM5AD3XCDBRAQIRTCMQN4P6","short_pith_number":"pith:LJ5DM5AD","schema_version":"1.0","canonical_sha256":"5a7a367403ddc430c41044662641bc7f80795057724f8d272fe31bc176610859","source":{"kind":"arxiv","id":"2602.06886","version":3},"attestation_state":"computed","paper":{"title":"Prompt Reinjection: Alleviating Prompt Forgetting in Multimodal Diffusion Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hui Li, Jingdong Wang, Kaihui Cheng, Qipeng Guo, Siyu Zhu, Yuwei Sun, Yuxuan Chen, Yuxuan Yao, Zilong Dong","submitted_at":"2026-02-06T17:19:53Z","abstract_excerpt":"Multimodal Diffusion Transformers (MMDiTs) for text-to-image generation maintain separate text and image branches, with bidirectional information flow between text tokens and visual latents throughout denoising. In this setting, we observe a prompt forgetting phenomenon: the semantics of the prompt representation in the text branch is progressively forgotten as depth increases. We further verify this effect on three representative MMDiTs--SD3, SD3.5, and FLUX.1 by probing linguistic attributes of the representations over the layers in the text branch. Motivated by these findings, we introduce "},"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":"2602.06886","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-02-06T17:19:53Z","cross_cats_sorted":[],"title_canon_sha256":"93786a1f2dd83841f35a5e7f4c158ca76c5692ac5b9be0aaee0b35d0e1381e23","abstract_canon_sha256":"9696ba539d6c79077e99bf630438cf98394215e5fcf337004f7edf6b100dbab0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:04:22.465502Z","signature_b64":"OoqtrZMV6r6YSqs4rfdwQUaH9uWaqLwcBmW76nqo3ew8jmU1BC1/QBVjJKx9n+fW4xcsunbuXaQNtdcqziUZDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a7a367403ddc430c41044662641bc7f80795057724f8d272fe31bc176610859","last_reissued_at":"2026-05-21T01:04:22.464562Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:04:22.464562Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Prompt Reinjection: Alleviating Prompt Forgetting in Multimodal Diffusion Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hui Li, Jingdong Wang, Kaihui Cheng, Qipeng Guo, Siyu Zhu, Yuwei Sun, Yuxuan Chen, Yuxuan Yao, Zilong Dong","submitted_at":"2026-02-06T17:19:53Z","abstract_excerpt":"Multimodal Diffusion Transformers (MMDiTs) for text-to-image generation maintain separate text and image branches, with bidirectional information flow between text tokens and visual latents throughout denoising. In this setting, we observe a prompt forgetting phenomenon: the semantics of the prompt representation in the text branch is progressively forgotten as depth increases. We further verify this effect on three representative MMDiTs--SD3, SD3.5, and FLUX.1 by probing linguistic attributes of the representations over the layers in the text branch. Motivated by these findings, we introduce "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.06886","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/2602.06886/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":"2602.06886","created_at":"2026-05-21T01:04:22.464689+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.06886v3","created_at":"2026-05-21T01:04:22.464689+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.06886","created_at":"2026-05-21T01:04:22.464689+00:00"},{"alias_kind":"pith_short_12","alias_value":"LJ5DM5AD3XCD","created_at":"2026-05-21T01:04:22.464689+00:00"},{"alias_kind":"pith_short_16","alias_value":"LJ5DM5AD3XCDBRAQ","created_at":"2026-05-21T01:04:22.464689+00:00"},{"alias_kind":"pith_short_8","alias_value":"LJ5DM5AD","created_at":"2026-05-21T01:04:22.464689+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LJ5DM5AD3XCDBRAQIRTCMQN4P6","json":"https://pith.science/pith/LJ5DM5AD3XCDBRAQIRTCMQN4P6.json","graph_json":"https://pith.science/api/pith-number/LJ5DM5AD3XCDBRAQIRTCMQN4P6/graph.json","events_json":"https://pith.science/api/pith-number/LJ5DM5AD3XCDBRAQIRTCMQN4P6/events.json","paper":"https://pith.science/paper/LJ5DM5AD"},"agent_actions":{"view_html":"https://pith.science/pith/LJ5DM5AD3XCDBRAQIRTCMQN4P6","download_json":"https://pith.science/pith/LJ5DM5AD3XCDBRAQIRTCMQN4P6.json","view_paper":"https://pith.science/paper/LJ5DM5AD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.06886&json=true","fetch_graph":"https://pith.science/api/pith-number/LJ5DM5AD3XCDBRAQIRTCMQN4P6/graph.json","fetch_events":"https://pith.science/api/pith-number/LJ5DM5AD3XCDBRAQIRTCMQN4P6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LJ5DM5AD3XCDBRAQIRTCMQN4P6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LJ5DM5AD3XCDBRAQIRTCMQN4P6/action/storage_attestation","attest_author":"https://pith.science/pith/LJ5DM5AD3XCDBRAQIRTCMQN4P6/action/author_attestation","sign_citation":"https://pith.science/pith/LJ5DM5AD3XCDBRAQIRTCMQN4P6/action/citation_signature","submit_replication":"https://pith.science/pith/LJ5DM5AD3XCDBRAQIRTCMQN4P6/action/replication_record"}},"created_at":"2026-05-21T01:04:22.464689+00:00","updated_at":"2026-05-21T01:04:22.464689+00:00"}