{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:J3ZI2F4ZSARQXC7ZNWGYRIFIVH","short_pith_number":"pith:J3ZI2F4Z","canonical_record":{"source":{"id":"2302.12066","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-02-23T14:43:53Z","cross_cats_sorted":[],"title_canon_sha256":"52917276e3f7dfe71758512706c4de3789427c6c399b9af30e7fb9c6b2d220ff","abstract_canon_sha256":"39710801298224c87463e758aa80b0fd383e896576b2af1ca986ee933ae889b1"},"schema_version":"1.0"},"canonical_sha256":"4ef28d179990230b8bf96d8d88a0a8a9e2c305338cda36938e31f18a1aeaa736","source":{"kind":"arxiv","id":"2302.12066","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.12066","created_at":"2026-07-05T05:44:59Z"},{"alias_kind":"arxiv_version","alias_value":"2302.12066v1","created_at":"2026-07-05T05:44:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.12066","created_at":"2026-07-05T05:44:59Z"},{"alias_kind":"pith_short_12","alias_value":"J3ZI2F4ZSARQ","created_at":"2026-07-05T05:44:59Z"},{"alias_kind":"pith_short_16","alias_value":"J3ZI2F4ZSARQXC7Z","created_at":"2026-07-05T05:44:59Z"},{"alias_kind":"pith_short_8","alias_value":"J3ZI2F4Z","created_at":"2026-07-05T05:44:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:J3ZI2F4ZSARQXC7ZNWGYRIFIVH","target":"record","payload":{"canonical_record":{"source":{"id":"2302.12066","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-02-23T14:43:53Z","cross_cats_sorted":[],"title_canon_sha256":"52917276e3f7dfe71758512706c4de3789427c6c399b9af30e7fb9c6b2d220ff","abstract_canon_sha256":"39710801298224c87463e758aa80b0fd383e896576b2af1ca986ee933ae889b1"},"schema_version":"1.0"},"canonical_sha256":"4ef28d179990230b8bf96d8d88a0a8a9e2c305338cda36938e31f18a1aeaa736","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:44:59.766987Z","signature_b64":"nNccjK8KhGS02MFhCXLcnUdMXBLMr0OPZlLL12ZPMg9+nP2avPbgJDAt6q75JtWYEM3HvzWQxx7Cb3ahZRzcDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4ef28d179990230b8bf96d8d88a0a8a9e2c305338cda36938e31f18a1aeaa736","last_reissued_at":"2026-07-05T05:44:59.766336Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:44:59.766336Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2302.12066","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:44:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wHY6KXuTYgsUFmSEDhIx/iRzR8JXOnps1qIjBqCNB80X/ERs0JG76mWYS7fvmiHt4jWjG3H7gl7yLzcXzZmhBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T09:40:56.331774Z"},"content_sha256":"bd737533e8b2484a4abb49eb92f24cdbdab3726d348c73acd68ea79e2694e26c","schema_version":"1.0","event_id":"sha256:bd737533e8b2484a4abb49eb92f24cdbdab3726d348c73acd68ea79e2694e26c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:J3ZI2F4ZSARQXC7ZNWGYRIFIVH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Teaching CLIP to Count to Ten","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ariel Ephrat, Inbar Mosseri, Michal Irani, Omer Tov, Roni Paiss, Shiran Zada, Tali Dekel","submitted_at":"2023-02-23T14:43:53Z","abstract_excerpt":"Large vision-language models (VLMs), such as CLIP, learn rich joint image-text representations, facilitating advances in numerous downstream tasks, including zero-shot classification and text-to-image generation. Nevertheless, existing VLMs exhibit a prominent well-documented limitation - they fail to encapsulate compositional concepts such as counting. We introduce a simple yet effective method to improve the quantitative understanding of VLMs, while maintaining their overall performance on common benchmarks. Specifically, we propose a new counting-contrastive loss used to finetune a pre-trai"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.12066","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/2302.12066/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":"2026-07-05T05:44:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GqJbDK5Us6QCGXQA5ofZr/n48mtt+qOWH3vmvfFp9Ra77AXgjhNVaLaurI5l/9lnd3gqqEzBcqoXM8xJ+1XwBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T09:40:56.332142Z"},"content_sha256":"5b35915e56c08c75e807b7f88cda105bb247c1b50019a4c34ed68176bc11ef9d","schema_version":"1.0","event_id":"sha256:5b35915e56c08c75e807b7f88cda105bb247c1b50019a4c34ed68176bc11ef9d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/J3ZI2F4ZSARQXC7ZNWGYRIFIVH/bundle.json","state_url":"https://pith.science/pith/J3ZI2F4ZSARQXC7ZNWGYRIFIVH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/J3ZI2F4ZSARQXC7ZNWGYRIFIVH/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":"2026-07-07T09:40:56Z","links":{"resolver":"https://pith.science/pith/J3ZI2F4ZSARQXC7ZNWGYRIFIVH","bundle":"https://pith.science/pith/J3ZI2F4ZSARQXC7ZNWGYRIFIVH/bundle.json","state":"https://pith.science/pith/J3ZI2F4ZSARQXC7ZNWGYRIFIVH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/J3ZI2F4ZSARQXC7ZNWGYRIFIVH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:J3ZI2F4ZSARQXC7ZNWGYRIFIVH","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":"39710801298224c87463e758aa80b0fd383e896576b2af1ca986ee933ae889b1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-02-23T14:43:53Z","title_canon_sha256":"52917276e3f7dfe71758512706c4de3789427c6c399b9af30e7fb9c6b2d220ff"},"schema_version":"1.0","source":{"id":"2302.12066","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.12066","created_at":"2026-07-05T05:44:59Z"},{"alias_kind":"arxiv_version","alias_value":"2302.12066v1","created_at":"2026-07-05T05:44:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.12066","created_at":"2026-07-05T05:44:59Z"},{"alias_kind":"pith_short_12","alias_value":"J3ZI2F4ZSARQ","created_at":"2026-07-05T05:44:59Z"},{"alias_kind":"pith_short_16","alias_value":"J3ZI2F4ZSARQXC7Z","created_at":"2026-07-05T05:44:59Z"},{"alias_kind":"pith_short_8","alias_value":"J3ZI2F4Z","created_at":"2026-07-05T05:44:59Z"}],"graph_snapshots":[{"event_id":"sha256:5b35915e56c08c75e807b7f88cda105bb247c1b50019a4c34ed68176bc11ef9d","target":"graph","created_at":"2026-07-05T05:44:59Z","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/2302.12066/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large vision-language models (VLMs), such as CLIP, learn rich joint image-text representations, facilitating advances in numerous downstream tasks, including zero-shot classification and text-to-image generation. Nevertheless, existing VLMs exhibit a prominent well-documented limitation - they fail to encapsulate compositional concepts such as counting. We introduce a simple yet effective method to improve the quantitative understanding of VLMs, while maintaining their overall performance on common benchmarks. Specifically, we propose a new counting-contrastive loss used to finetune a pre-trai","authors_text":"Ariel Ephrat, Inbar Mosseri, Michal Irani, Omer Tov, Roni Paiss, Shiran Zada, Tali Dekel","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-02-23T14:43:53Z","title":"Teaching CLIP to Count to Ten"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.12066","kind":"arxiv","version":1},"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:bd737533e8b2484a4abb49eb92f24cdbdab3726d348c73acd68ea79e2694e26c","target":"record","created_at":"2026-07-05T05:44:59Z","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":"39710801298224c87463e758aa80b0fd383e896576b2af1ca986ee933ae889b1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-02-23T14:43:53Z","title_canon_sha256":"52917276e3f7dfe71758512706c4de3789427c6c399b9af30e7fb9c6b2d220ff"},"schema_version":"1.0","source":{"id":"2302.12066","kind":"arxiv","version":1}},"canonical_sha256":"4ef28d179990230b8bf96d8d88a0a8a9e2c305338cda36938e31f18a1aeaa736","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4ef28d179990230b8bf96d8d88a0a8a9e2c305338cda36938e31f18a1aeaa736","first_computed_at":"2026-07-05T05:44:59.766336Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:44:59.766336Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nNccjK8KhGS02MFhCXLcnUdMXBLMr0OPZlLL12ZPMg9+nP2avPbgJDAt6q75JtWYEM3HvzWQxx7Cb3ahZRzcDA==","signature_status":"signed_v1","signed_at":"2026-07-05T05:44:59.766987Z","signed_message":"canonical_sha256_bytes"},"source_id":"2302.12066","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bd737533e8b2484a4abb49eb92f24cdbdab3726d348c73acd68ea79e2694e26c","sha256:5b35915e56c08c75e807b7f88cda105bb247c1b50019a4c34ed68176bc11ef9d"],"state_sha256":"319bbf79eb50453115f43266abdfe9c1b085bd96fa98c81d620171c0c658115a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8TcUXcMMG3qJnXl5HaE4Tz6DBMrHLU4eNWWp+qsJc7nMC1/CvhFZHbyx/qOUtvt+ImdjXrf9cRRrRR2wXTzfDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T09:40:56.334085Z","bundle_sha256":"dce32cda0570bb0a18061a1ecd122b9084192a7197686867319bf1e17f596f9d"}}