{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HY6L4YSJDWUSQRRHF2ULSQB2HU","short_pith_number":"pith:HY6L4YSJ","schema_version":"1.0","canonical_sha256":"3e3cbe62491da92846272ea8b9403a3d3a7e4c3fb4812ff5b4bdac57b310f4dd","source":{"kind":"arxiv","id":"2605.21907","version":1},"attestation_state":"computed","paper":{"title":"Guided Trajectory Optimization with Sparse Scaling for Test-Time Diffusion","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gang Dai, Guohao Chen, Shuaicheng Niu, YiMing Xia, Yining Huang","submitted_at":"2026-05-21T02:30:01Z","abstract_excerpt":"The efficient Test-Time Scaling (TTS) paradigm offers a promising perspective for enhancing the generation performance of diffusion models. However, current solutions are limited to a static, pre-defined noise pool and suffer from inflexible noise exploration across the denoising trajectory. To bridge this gap, we propose RTS, a novel Reward-guided Trajectory Scaling method to fully unlock the generative potential of diffusion models. Unlike existing methods, RTS facilitates the synthesis of refined, high-fidelity images via two core innovations: 1) a reward-guided noise optimization strategy "},"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":"2605.21907","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-21T02:30:01Z","cross_cats_sorted":[],"title_canon_sha256":"ff083df0d248004f5ccb68a925788b86efe96622e17e8776920a66e4f50cb82a","abstract_canon_sha256":"2a66856143536975bcc316075121a7d6dbbdb732b9711d1f70f9ef400d105e58"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:04:13.961606Z","signature_b64":"Y/ViVP3ldw2GAXJNQqC6r8nRPk2N+sSh+Tt+Nenc+hmfiDPK99VKT6XhWeteVIUdGeCLKM7YngG27qK+W79pBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3e3cbe62491da92846272ea8b9403a3d3a7e4c3fb4812ff5b4bdac57b310f4dd","last_reissued_at":"2026-05-22T01:04:13.960843Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:04:13.960843Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Guided Trajectory Optimization with Sparse Scaling for Test-Time Diffusion","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gang Dai, Guohao Chen, Shuaicheng Niu, YiMing Xia, Yining Huang","submitted_at":"2026-05-21T02:30:01Z","abstract_excerpt":"The efficient Test-Time Scaling (TTS) paradigm offers a promising perspective for enhancing the generation performance of diffusion models. However, current solutions are limited to a static, pre-defined noise pool and suffer from inflexible noise exploration across the denoising trajectory. To bridge this gap, we propose RTS, a novel Reward-guided Trajectory Scaling method to fully unlock the generative potential of diffusion models. Unlike existing methods, RTS facilitates the synthesis of refined, high-fidelity images via two core innovations: 1) a reward-guided noise optimization strategy "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21907","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/2605.21907/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":"2605.21907","created_at":"2026-05-22T01:04:13.960951+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.21907v1","created_at":"2026-05-22T01:04:13.960951+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21907","created_at":"2026-05-22T01:04:13.960951+00:00"},{"alias_kind":"pith_short_12","alias_value":"HY6L4YSJDWUS","created_at":"2026-05-22T01:04:13.960951+00:00"},{"alias_kind":"pith_short_16","alias_value":"HY6L4YSJDWUSQRRH","created_at":"2026-05-22T01:04:13.960951+00:00"},{"alias_kind":"pith_short_8","alias_value":"HY6L4YSJ","created_at":"2026-05-22T01:04:13.960951+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/HY6L4YSJDWUSQRRHF2ULSQB2HU","json":"https://pith.science/pith/HY6L4YSJDWUSQRRHF2ULSQB2HU.json","graph_json":"https://pith.science/api/pith-number/HY6L4YSJDWUSQRRHF2ULSQB2HU/graph.json","events_json":"https://pith.science/api/pith-number/HY6L4YSJDWUSQRRHF2ULSQB2HU/events.json","paper":"https://pith.science/paper/HY6L4YSJ"},"agent_actions":{"view_html":"https://pith.science/pith/HY6L4YSJDWUSQRRHF2ULSQB2HU","download_json":"https://pith.science/pith/HY6L4YSJDWUSQRRHF2ULSQB2HU.json","view_paper":"https://pith.science/paper/HY6L4YSJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.21907&json=true","fetch_graph":"https://pith.science/api/pith-number/HY6L4YSJDWUSQRRHF2ULSQB2HU/graph.json","fetch_events":"https://pith.science/api/pith-number/HY6L4YSJDWUSQRRHF2ULSQB2HU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HY6L4YSJDWUSQRRHF2ULSQB2HU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HY6L4YSJDWUSQRRHF2ULSQB2HU/action/storage_attestation","attest_author":"https://pith.science/pith/HY6L4YSJDWUSQRRHF2ULSQB2HU/action/author_attestation","sign_citation":"https://pith.science/pith/HY6L4YSJDWUSQRRHF2ULSQB2HU/action/citation_signature","submit_replication":"https://pith.science/pith/HY6L4YSJDWUSQRRHF2ULSQB2HU/action/replication_record"}},"created_at":"2026-05-22T01:04:13.960951+00:00","updated_at":"2026-05-22T01:04:13.960951+00:00"}