{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:AINAOZI5YHIQR7WOSBZDPIDQ3Z","short_pith_number":"pith:AINAOZI5","canonical_record":{"source":{"id":"2412.11917","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-16T16:01:18Z","cross_cats_sorted":[],"title_canon_sha256":"489713e8dbe350ff9586b12752d10d6b84406ff9d565d7cb3354d10c9822cdd6","abstract_canon_sha256":"8432e569003e463eefc4d3878a04ef3fc24f6b16646ce0c508f6ccdb8af71004"},"schema_version":"1.0"},"canonical_sha256":"021a07651dc1d108fece907237a070de7eb53b07f82fd1b8de5567b730b785bc","source":{"kind":"arxiv","id":"2412.11917","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.11917","created_at":"2026-07-05T09:51:40Z"},{"alias_kind":"arxiv_version","alias_value":"2412.11917v3","created_at":"2026-07-05T09:51:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.11917","created_at":"2026-07-05T09:51:40Z"},{"alias_kind":"pith_short_12","alias_value":"AINAOZI5YHIQ","created_at":"2026-07-05T09:51:40Z"},{"alias_kind":"pith_short_16","alias_value":"AINAOZI5YHIQR7WO","created_at":"2026-07-05T09:51:40Z"},{"alias_kind":"pith_short_8","alias_value":"AINAOZI5","created_at":"2026-07-05T09:51:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:AINAOZI5YHIQR7WOSBZDPIDQ3Z","target":"record","payload":{"canonical_record":{"source":{"id":"2412.11917","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-16T16:01:18Z","cross_cats_sorted":[],"title_canon_sha256":"489713e8dbe350ff9586b12752d10d6b84406ff9d565d7cb3354d10c9822cdd6","abstract_canon_sha256":"8432e569003e463eefc4d3878a04ef3fc24f6b16646ce0c508f6ccdb8af71004"},"schema_version":"1.0"},"canonical_sha256":"021a07651dc1d108fece907237a070de7eb53b07f82fd1b8de5567b730b785bc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:51:40.071035Z","signature_b64":"xAogQHWIrFQGlbLB3EcaHsgfOPW3f3pcf/3eniv5ENjgppD+rvCQNd1oCLZz+BGnOTOtMXUKt8mpBCnKnUp8BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"021a07651dc1d108fece907237a070de7eb53b07f82fd1b8de5567b730b785bc","last_reissued_at":"2026-07-05T09:51:40.070569Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:51:40.070569Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.11917","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:51:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3qKkn+XPMIsZVmXC76OPfl7cGuB9pKqJpcheFtS0GuquFzpxG0nTAJlrQPEl1+AQuHWnvyr0sIGh0LYpdCASDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T08:32:51.144424Z"},"content_sha256":"bbb97ae3cfded051a0fe2a4c89a9c1fa90571a52f19a1dfea819533f66e255fe","schema_version":"1.0","event_id":"sha256:bbb97ae3cfded051a0fe2a4c89a9c1fa90571a52f19a1dfea819533f66e255fe"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:AINAOZI5YHIQR7WOSBZDPIDQ3Z","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Does VLM Classification Benefit from LLM Description Semantics?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bj\\\"orn Ommer, Dmytro Kotovenko, Lennart Rietdorf, Pingchuan Ma, Vincent Tao Hu","submitted_at":"2024-12-16T16:01:18Z","abstract_excerpt":"Accurately describing images with text is a foundation of explainable AI. Vision-Language Models (VLMs) like CLIP have recently addressed this by aligning images and texts in a shared embedding space, expressing semantic similarities between vision and language embeddings. VLM classification can be improved with descriptions generated by Large Language Models (LLMs). However, it is difficult to determine the contribution of actual description semantics, as the performance gain may also stem from a semantic-agnostic ensembling effect, where multiple modified text prompts act as a noisy test-tim"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.11917","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/2412.11917/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":"2026-07-05T09:51:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pvUKuFd6GmFDeM4Ed+TiygpW6k73PVgm+z4S3VH7R2iSp4A7yRe/Js5TH2MEv0HcpObAwLJZCM1dRzUPj7g9AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T08:32:51.144996Z"},"content_sha256":"4bedc7a53b4c02d4937d632aa6b61b6e6ad55aeb8b1c4bed6c1f9d6efea7b444","schema_version":"1.0","event_id":"sha256:4bedc7a53b4c02d4937d632aa6b61b6e6ad55aeb8b1c4bed6c1f9d6efea7b444"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AINAOZI5YHIQR7WOSBZDPIDQ3Z/bundle.json","state_url":"https://pith.science/pith/AINAOZI5YHIQR7WOSBZDPIDQ3Z/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AINAOZI5YHIQR7WOSBZDPIDQ3Z/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":"2026-08-12T08:32:51Z","links":{"resolver":"https://pith.science/pith/AINAOZI5YHIQR7WOSBZDPIDQ3Z","bundle":"https://pith.science/pith/AINAOZI5YHIQR7WOSBZDPIDQ3Z/bundle.json","state":"https://pith.science/pith/AINAOZI5YHIQR7WOSBZDPIDQ3Z/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AINAOZI5YHIQR7WOSBZDPIDQ3Z/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:AINAOZI5YHIQR7WOSBZDPIDQ3Z","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":"8432e569003e463eefc4d3878a04ef3fc24f6b16646ce0c508f6ccdb8af71004","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-16T16:01:18Z","title_canon_sha256":"489713e8dbe350ff9586b12752d10d6b84406ff9d565d7cb3354d10c9822cdd6"},"schema_version":"1.0","source":{"id":"2412.11917","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.11917","created_at":"2026-07-05T09:51:40Z"},{"alias_kind":"arxiv_version","alias_value":"2412.11917v3","created_at":"2026-07-05T09:51:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.11917","created_at":"2026-07-05T09:51:40Z"},{"alias_kind":"pith_short_12","alias_value":"AINAOZI5YHIQ","created_at":"2026-07-05T09:51:40Z"},{"alias_kind":"pith_short_16","alias_value":"AINAOZI5YHIQR7WO","created_at":"2026-07-05T09:51:40Z"},{"alias_kind":"pith_short_8","alias_value":"AINAOZI5","created_at":"2026-07-05T09:51:40Z"}],"graph_snapshots":[{"event_id":"sha256:4bedc7a53b4c02d4937d632aa6b61b6e6ad55aeb8b1c4bed6c1f9d6efea7b444","target":"graph","created_at":"2026-07-05T09:51:40Z","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/2412.11917/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Accurately describing images with text is a foundation of explainable AI. Vision-Language Models (VLMs) like CLIP have recently addressed this by aligning images and texts in a shared embedding space, expressing semantic similarities between vision and language embeddings. VLM classification can be improved with descriptions generated by Large Language Models (LLMs). However, it is difficult to determine the contribution of actual description semantics, as the performance gain may also stem from a semantic-agnostic ensembling effect, where multiple modified text prompts act as a noisy test-tim","authors_text":"Bj\\\"orn Ommer, Dmytro Kotovenko, Lennart Rietdorf, Pingchuan Ma, Vincent Tao Hu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-16T16:01:18Z","title":"Does VLM Classification Benefit from LLM Description Semantics?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.11917","kind":"arxiv","version":3},"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:bbb97ae3cfded051a0fe2a4c89a9c1fa90571a52f19a1dfea819533f66e255fe","target":"record","created_at":"2026-07-05T09:51:40Z","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":"8432e569003e463eefc4d3878a04ef3fc24f6b16646ce0c508f6ccdb8af71004","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-16T16:01:18Z","title_canon_sha256":"489713e8dbe350ff9586b12752d10d6b84406ff9d565d7cb3354d10c9822cdd6"},"schema_version":"1.0","source":{"id":"2412.11917","kind":"arxiv","version":3}},"canonical_sha256":"021a07651dc1d108fece907237a070de7eb53b07f82fd1b8de5567b730b785bc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"021a07651dc1d108fece907237a070de7eb53b07f82fd1b8de5567b730b785bc","first_computed_at":"2026-07-05T09:51:40.070569Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:51:40.070569Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xAogQHWIrFQGlbLB3EcaHsgfOPW3f3pcf/3eniv5ENjgppD+rvCQNd1oCLZz+BGnOTOtMXUKt8mpBCnKnUp8BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:51:40.071035Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.11917","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bbb97ae3cfded051a0fe2a4c89a9c1fa90571a52f19a1dfea819533f66e255fe","sha256:4bedc7a53b4c02d4937d632aa6b61b6e6ad55aeb8b1c4bed6c1f9d6efea7b444"],"state_sha256":"d289ea13af50cfc701e129cb4f368b38314a5ff0899c782e6cd191228d3d481a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"J5eS7cCkSTlibvfET5vrB7nSzgRYNXg2PcEOeKNxtWmyXS0C+4cpqfaUeSkCYmkCj2Kh3tbb28fB8ABeS09TAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T08:32:51.206630Z","bundle_sha256":"a38ecc9887d38f96f730682e979b82e6cc54ac1ba7db7b5e56240dfd0ad92405"}}