{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:FOP2LENPCGACZPG7KH27YHNAON","short_pith_number":"pith:FOP2LENP","schema_version":"1.0","canonical_sha256":"2b9fa591af11802cbcdf51f5fc1da0736abaa67e654b5aff06c753cc72889f63","source":{"kind":"arxiv","id":"2312.14733","version":1},"attestation_state":"computed","paper":{"title":"Harnessing Diffusion Models for Visual Perception with Meta Prompts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bingyi Kang, Jiashi Feng, Li Zhang, Qiang Wan, Zilong Huang","submitted_at":"2023-12-22T14:40:55Z","abstract_excerpt":"The issue of generative pretraining for vision models has persisted as a long-standing conundrum. At present, the text-to-image (T2I) diffusion model demonstrates remarkable proficiency in generating high-definition images matching textual inputs, a feat made possible through its pre-training on large-scale image-text pairs. This leads to a natural inquiry: can diffusion models be utilized to tackle visual perception tasks? In this paper, we propose a simple yet effective scheme to harness a diffusion model for visual perception tasks. Our key insight is to introduce learnable embeddings (meta"},"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":"2312.14733","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-12-22T14:40:55Z","cross_cats_sorted":[],"title_canon_sha256":"e7b6629c929eaa52fa4a1ecfc7f73ab54fa0caf5b3745cd87394f89f22a815b1","abstract_canon_sha256":"e623dee918b0be093e6c8b29a3e3e35d00e59c5779db765f3361d1eb531cc9b1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:27:19.136752Z","signature_b64":"Gu39/AnHjWH8T1YnP5ihQzwD0PBUA4Wez+rANCUll0C3xlYC59bRGp4LBX7o9hTUPLqXONcWpb6LUZ7VXmjWBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2b9fa591af11802cbcdf51f5fc1da0736abaa67e654b5aff06c753cc72889f63","last_reissued_at":"2026-07-05T07:27:19.136329Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:27:19.136329Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Harnessing Diffusion Models for Visual Perception with Meta Prompts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bingyi Kang, Jiashi Feng, Li Zhang, Qiang Wan, Zilong Huang","submitted_at":"2023-12-22T14:40:55Z","abstract_excerpt":"The issue of generative pretraining for vision models has persisted as a long-standing conundrum. At present, the text-to-image (T2I) diffusion model demonstrates remarkable proficiency in generating high-definition images matching textual inputs, a feat made possible through its pre-training on large-scale image-text pairs. This leads to a natural inquiry: can diffusion models be utilized to tackle visual perception tasks? In this paper, we propose a simple yet effective scheme to harness a diffusion model for visual perception tasks. Our key insight is to introduce learnable embeddings (meta"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.14733","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/2312.14733/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":"2312.14733","created_at":"2026-07-05T07:27:19.136390+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.14733v1","created_at":"2026-07-05T07:27:19.136390+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.14733","created_at":"2026-07-05T07:27:19.136390+00:00"},{"alias_kind":"pith_short_12","alias_value":"FOP2LENPCGAC","created_at":"2026-07-05T07:27:19.136390+00:00"},{"alias_kind":"pith_short_16","alias_value":"FOP2LENPCGACZPG7","created_at":"2026-07-05T07:27:19.136390+00:00"},{"alias_kind":"pith_short_8","alias_value":"FOP2LENP","created_at":"2026-07-05T07:27:19.136390+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2504.18190","citing_title":"What is the Added Value of UDA in the VFM Era?","ref_index":50,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FOP2LENPCGACZPG7KH27YHNAON","json":"https://pith.science/pith/FOP2LENPCGACZPG7KH27YHNAON.json","graph_json":"https://pith.science/api/pith-number/FOP2LENPCGACZPG7KH27YHNAON/graph.json","events_json":"https://pith.science/api/pith-number/FOP2LENPCGACZPG7KH27YHNAON/events.json","paper":"https://pith.science/paper/FOP2LENP"},"agent_actions":{"view_html":"https://pith.science/pith/FOP2LENPCGACZPG7KH27YHNAON","download_json":"https://pith.science/pith/FOP2LENPCGACZPG7KH27YHNAON.json","view_paper":"https://pith.science/paper/FOP2LENP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.14733&json=true","fetch_graph":"https://pith.science/api/pith-number/FOP2LENPCGACZPG7KH27YHNAON/graph.json","fetch_events":"https://pith.science/api/pith-number/FOP2LENPCGACZPG7KH27YHNAON/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FOP2LENPCGACZPG7KH27YHNAON/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FOP2LENPCGACZPG7KH27YHNAON/action/storage_attestation","attest_author":"https://pith.science/pith/FOP2LENPCGACZPG7KH27YHNAON/action/author_attestation","sign_citation":"https://pith.science/pith/FOP2LENPCGACZPG7KH27YHNAON/action/citation_signature","submit_replication":"https://pith.science/pith/FOP2LENPCGACZPG7KH27YHNAON/action/replication_record"}},"created_at":"2026-07-05T07:27:19.136390+00:00","updated_at":"2026-07-05T07:27:19.136390+00:00"}