{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:D6ZBCV4XJEQTSRWYSXZNM3EFPA","short_pith_number":"pith:D6ZBCV4X","canonical_record":{"source":{"id":"2510.25897","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-10-29T18:59:17Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"650656a4d394d05b60bdc4412643e109b2d1a1ed1171f0d60d0a78853ac261c9","abstract_canon_sha256":"9d0a7f0cec732c5ac0ef17471aa0cf4b810d10a924dda1b05f4a7e1a6753de61"},"schema_version":"1.0"},"canonical_sha256":"1fb211579749213946d895f2d66c857838fab47010f19752803415bfd1a5dc61","source":{"kind":"arxiv","id":"2510.25897","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.25897","created_at":"2026-05-20T02:05:37Z"},{"alias_kind":"arxiv_version","alias_value":"2510.25897v2","created_at":"2026-05-20T02:05:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.25897","created_at":"2026-05-20T02:05:37Z"},{"alias_kind":"pith_short_12","alias_value":"D6ZBCV4XJEQT","created_at":"2026-05-20T02:05:37Z"},{"alias_kind":"pith_short_16","alias_value":"D6ZBCV4XJEQTSRWY","created_at":"2026-05-20T02:05:37Z"},{"alias_kind":"pith_short_8","alias_value":"D6ZBCV4X","created_at":"2026-05-20T02:05:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:D6ZBCV4XJEQTSRWYSXZNM3EFPA","target":"record","payload":{"canonical_record":{"source":{"id":"2510.25897","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-10-29T18:59:17Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"650656a4d394d05b60bdc4412643e109b2d1a1ed1171f0d60d0a78853ac261c9","abstract_canon_sha256":"9d0a7f0cec732c5ac0ef17471aa0cf4b810d10a924dda1b05f4a7e1a6753de61"},"schema_version":"1.0"},"canonical_sha256":"1fb211579749213946d895f2d66c857838fab47010f19752803415bfd1a5dc61","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T02:05:37.763257Z","signature_b64":"W6P8x7+vr4TXQXVtqCiex/MOOgxPr8LV2WVUgp3Vmz60r6tkEkg7Yc4zyWmf89PVrN+VRvGwPoUAKQfV2eq6DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1fb211579749213946d895f2d66c857838fab47010f19752803415bfd1a5dc61","last_reissued_at":"2026-05-20T02:05:37.762384Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T02:05:37.762384Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2510.25897","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T02:05:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XDV7oi7Qbwfv6jHO+B3QR8BDJAWxoZ3ajEdw9gdqjwrPG9uC/ljRkYJnfYCfZ4zkdW2x7VjlCk8FdurFgJmRAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T00:10:33.916695Z"},"content_sha256":"8903a511a83baa2b00d5acd73daf962c06ffd5460a43a1be37ce5d165a18e257","schema_version":"1.0","event_id":"sha256:8903a511a83baa2b00d5acd73daf962c06ffd5460a43a1be37ce5d165a18e257"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:D6ZBCV4XJEQTSRWYSXZNM3EFPA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MIRO: MultI-Reward cOnditioned pretraining improves T2I quality and efficiency","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Arijit Ghosh, David Picard, Lucas Degeorge, Nicolas Dufour, Vicky Kalogeiton","submitted_at":"2025-10-29T18:59:17Z","abstract_excerpt":"The default paradigm of post-training text-to-image generators includes post-hoc selection of generated images, and subsequent training with one reward model to align the generator to the reward, typically user preference. This discards informative data as well as optimizes only for a single reward, hence harming diversity, semantic fidelity and efficiency. Instead, we propose MIRO, a method that conditions the model on multiple rewards during training, thus letting the model learn user preferences directly. MIRO pre-training both improves the visual quality of the generated images and speeds "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.25897","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/2510.25897/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T02:05:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A7l5eJT4lOUEYVpFtzT4I68/R5edJ2niVaEUlPXOcot9SM1oeYzAPairjTZn4DRnFKsuWwrmf7ACuOdyGpmcDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T00:10:33.917356Z"},"content_sha256":"93c4315c0cca049eb4c373477f50d17af3f3addfa11c8ec320c016918212c657","schema_version":"1.0","event_id":"sha256:93c4315c0cca049eb4c373477f50d17af3f3addfa11c8ec320c016918212c657"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/D6ZBCV4XJEQTSRWYSXZNM3EFPA/bundle.json","state_url":"https://pith.science/pith/D6ZBCV4XJEQTSRWYSXZNM3EFPA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/D6ZBCV4XJEQTSRWYSXZNM3EFPA/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-27T00:10:33Z","links":{"resolver":"https://pith.science/pith/D6ZBCV4XJEQTSRWYSXZNM3EFPA","bundle":"https://pith.science/pith/D6ZBCV4XJEQTSRWYSXZNM3EFPA/bundle.json","state":"https://pith.science/pith/D6ZBCV4XJEQTSRWYSXZNM3EFPA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/D6ZBCV4XJEQTSRWYSXZNM3EFPA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:D6ZBCV4XJEQTSRWYSXZNM3EFPA","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"9d0a7f0cec732c5ac0ef17471aa0cf4b810d10a924dda1b05f4a7e1a6753de61","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-10-29T18:59:17Z","title_canon_sha256":"650656a4d394d05b60bdc4412643e109b2d1a1ed1171f0d60d0a78853ac261c9"},"schema_version":"1.0","source":{"id":"2510.25897","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.25897","created_at":"2026-05-20T02:05:37Z"},{"alias_kind":"arxiv_version","alias_value":"2510.25897v2","created_at":"2026-05-20T02:05:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.25897","created_at":"2026-05-20T02:05:37Z"},{"alias_kind":"pith_short_12","alias_value":"D6ZBCV4XJEQT","created_at":"2026-05-20T02:05:37Z"},{"alias_kind":"pith_short_16","alias_value":"D6ZBCV4XJEQTSRWY","created_at":"2026-05-20T02:05:37Z"},{"alias_kind":"pith_short_8","alias_value":"D6ZBCV4X","created_at":"2026-05-20T02:05:37Z"}],"graph_snapshots":[{"event_id":"sha256:93c4315c0cca049eb4c373477f50d17af3f3addfa11c8ec320c016918212c657","target":"graph","created_at":"2026-05-20T02:05:37Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2510.25897/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The default paradigm of post-training text-to-image generators includes post-hoc selection of generated images, and subsequent training with one reward model to align the generator to the reward, typically user preference. This discards informative data as well as optimizes only for a single reward, hence harming diversity, semantic fidelity and efficiency. Instead, we propose MIRO, a method that conditions the model on multiple rewards during training, thus letting the model learn user preferences directly. MIRO pre-training both improves the visual quality of the generated images and speeds ","authors_text":"Arijit Ghosh, David Picard, Lucas Degeorge, Nicolas Dufour, Vicky Kalogeiton","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-10-29T18:59:17Z","title":"MIRO: MultI-Reward cOnditioned pretraining improves T2I quality and efficiency"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.25897","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8903a511a83baa2b00d5acd73daf962c06ffd5460a43a1be37ce5d165a18e257","target":"record","created_at":"2026-05-20T02:05:37Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"9d0a7f0cec732c5ac0ef17471aa0cf4b810d10a924dda1b05f4a7e1a6753de61","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-10-29T18:59:17Z","title_canon_sha256":"650656a4d394d05b60bdc4412643e109b2d1a1ed1171f0d60d0a78853ac261c9"},"schema_version":"1.0","source":{"id":"2510.25897","kind":"arxiv","version":2}},"canonical_sha256":"1fb211579749213946d895f2d66c857838fab47010f19752803415bfd1a5dc61","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1fb211579749213946d895f2d66c857838fab47010f19752803415bfd1a5dc61","first_computed_at":"2026-05-20T02:05:37.762384Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T02:05:37.762384Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"W6P8x7+vr4TXQXVtqCiex/MOOgxPr8LV2WVUgp3Vmz60r6tkEkg7Yc4zyWmf89PVrN+VRvGwPoUAKQfV2eq6DA==","signature_status":"signed_v1","signed_at":"2026-05-20T02:05:37.763257Z","signed_message":"canonical_sha256_bytes"},"source_id":"2510.25897","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8903a511a83baa2b00d5acd73daf962c06ffd5460a43a1be37ce5d165a18e257","sha256:93c4315c0cca049eb4c373477f50d17af3f3addfa11c8ec320c016918212c657"],"state_sha256":"889353adcbb600a0cf867af54bcd0901a1d66edb82a09ca542ab6bfe1d3b855f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JNtdPCkbGOxONKDBSBj9eeYyvuf0gCl+Fi+p+I/zVRkzCk+q2D4NGFxXAycIUiHglAnKw/aZHBwQcdCy1wgOBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T00:10:33.920937Z","bundle_sha256":"66c75eb1ae1630bb16e076753a74440b1262a99133bcd244eb36b582a3bed2cf"}}