{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KX4CIXDCWR376JWVJXQPD5ZKGZ","short_pith_number":"pith:KX4CIXDC","schema_version":"1.0","canonical_sha256":"55f8245c62b477ff26d54de0f1f72a367f417765bacbd0974f79157e31bb4c78","source":{"kind":"arxiv","id":"2506.02708","version":1},"attestation_state":"computed","paper":{"title":"Iterative Self-Improvement of Vision Language Models for Image Scoring and Self-Explanation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Naoto Tanji, Toshihiko Yamasaki","submitted_at":"2025-06-03T10:04:19Z","abstract_excerpt":"Image scoring is a crucial task in numerous real-world applications. To trust a model's judgment, understanding its rationale is essential. This paper proposes a novel training method for Vision Language Models (VLMs) to generate not only image scores but also corresponding justifications in natural language. Leveraging only an image scoring dataset and an instruction-tuned VLM, our method enables self-training, utilizing the VLM's generated text without relying on external data or models. In addition, we introduce a simple method for creating a dataset designed to improve alignment between pr"},"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":"2506.02708","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-03T10:04:19Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"132d665fcefbd44aab4885563b29491b8286f6476d557af984fec3f1a7262719","abstract_canon_sha256":"d2cfa791dfb7250b16bc451709fe3c0ac70a3eeea3ed75e8839cb6de9a48d065"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:15:06.886568Z","signature_b64":"8nCI8RDuvOdpI3b2hB7UxkcDj7FnPxL5Uk4hs9FVfQ0BSLFSVLXP8W19/DUF+uih+RXm8EphBw16ShHLjGHeAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"55f8245c62b477ff26d54de0f1f72a367f417765bacbd0974f79157e31bb4c78","last_reissued_at":"2026-07-05T11:15:06.886010Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:15:06.886010Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Iterative Self-Improvement of Vision Language Models for Image Scoring and Self-Explanation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Naoto Tanji, Toshihiko Yamasaki","submitted_at":"2025-06-03T10:04:19Z","abstract_excerpt":"Image scoring is a crucial task in numerous real-world applications. To trust a model's judgment, understanding its rationale is essential. This paper proposes a novel training method for Vision Language Models (VLMs) to generate not only image scores but also corresponding justifications in natural language. Leveraging only an image scoring dataset and an instruction-tuned VLM, our method enables self-training, utilizing the VLM's generated text without relying on external data or models. In addition, we introduce a simple method for creating a dataset designed to improve alignment between pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.02708","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/2506.02708/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":"2506.02708","created_at":"2026-07-05T11:15:06.886066+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.02708v1","created_at":"2026-07-05T11:15:06.886066+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.02708","created_at":"2026-07-05T11:15:06.886066+00:00"},{"alias_kind":"pith_short_12","alias_value":"KX4CIXDCWR37","created_at":"2026-07-05T11:15:06.886066+00:00"},{"alias_kind":"pith_short_16","alias_value":"KX4CIXDCWR376JWV","created_at":"2026-07-05T11:15:06.886066+00:00"},{"alias_kind":"pith_short_8","alias_value":"KX4CIXDC","created_at":"2026-07-05T11:15:06.886066+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.02708","citing_title":"Iterative Self-Improvement of Vision Language Models for Image Scoring and Self-Explanation","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KX4CIXDCWR376JWVJXQPD5ZKGZ","json":"https://pith.science/pith/KX4CIXDCWR376JWVJXQPD5ZKGZ.json","graph_json":"https://pith.science/api/pith-number/KX4CIXDCWR376JWVJXQPD5ZKGZ/graph.json","events_json":"https://pith.science/api/pith-number/KX4CIXDCWR376JWVJXQPD5ZKGZ/events.json","paper":"https://pith.science/paper/KX4CIXDC"},"agent_actions":{"view_html":"https://pith.science/pith/KX4CIXDCWR376JWVJXQPD5ZKGZ","download_json":"https://pith.science/pith/KX4CIXDCWR376JWVJXQPD5ZKGZ.json","view_paper":"https://pith.science/paper/KX4CIXDC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.02708&json=true","fetch_graph":"https://pith.science/api/pith-number/KX4CIXDCWR376JWVJXQPD5ZKGZ/graph.json","fetch_events":"https://pith.science/api/pith-number/KX4CIXDCWR376JWVJXQPD5ZKGZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KX4CIXDCWR376JWVJXQPD5ZKGZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KX4CIXDCWR376JWVJXQPD5ZKGZ/action/storage_attestation","attest_author":"https://pith.science/pith/KX4CIXDCWR376JWVJXQPD5ZKGZ/action/author_attestation","sign_citation":"https://pith.science/pith/KX4CIXDCWR376JWVJXQPD5ZKGZ/action/citation_signature","submit_replication":"https://pith.science/pith/KX4CIXDCWR376JWVJXQPD5ZKGZ/action/replication_record"}},"created_at":"2026-07-05T11:15:06.886066+00:00","updated_at":"2026-07-05T11:15:06.886066+00:00"}