{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:RRDKF72LEG4VWBCL373AKUHUER","short_pith_number":"pith:RRDKF72L","schema_version":"1.0","canonical_sha256":"8c46a2ff4b21b95b044bdff60550f4247c26c4ded55346fa865a8b81f6de8566","source":{"kind":"arxiv","id":"2306.07282","version":2},"attestation_state":"computed","paper":{"title":"Waffling around for Performance: Visual Classification with Random Words and Broad Concepts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"A. Sophia Koepke, Cordelia Schmid, Jae Myung Kim, Karsten Roth, Oriol Vinyals, Zeynep Akata","submitted_at":"2023-06-12T17:59:48Z","abstract_excerpt":"The visual classification performance of vision-language models such as CLIP has been shown to benefit from additional semantic knowledge from large language models (LLMs) such as GPT-3. In particular, averaging over LLM-generated class descriptors, e.g. \"waffle, which has a round shape\", can notably improve generalization performance. In this work, we critically study this behavior and propose WaffleCLIP, a framework for zero-shot visual classification which simply replaces LLM-generated descriptors with random character and word descriptors. Without querying external models, we achieve compa"},"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":"2306.07282","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-06-12T17:59:48Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"eb599ff300543b423359b45af5309212085d698a5c9b9a0878e152f303a7e189","abstract_canon_sha256":"ee0fe87b2fe967d7aa03f7605b0c5f16822e7986f0b2af5e2af12323b42bec95"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:42:06.334948Z","signature_b64":"vhcn4oSXqw5FfcDPvZG5Eu7MAROHaqUa3My5y/1c98+qbVH1kRQOyQJ64OtOBBZqjWVmfFOWuuOG9D5MBn8fDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8c46a2ff4b21b95b044bdff60550f4247c26c4ded55346fa865a8b81f6de8566","last_reissued_at":"2026-07-05T06:42:06.334210Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:42:06.334210Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Waffling around for Performance: Visual Classification with Random Words and Broad Concepts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"A. Sophia Koepke, Cordelia Schmid, Jae Myung Kim, Karsten Roth, Oriol Vinyals, Zeynep Akata","submitted_at":"2023-06-12T17:59:48Z","abstract_excerpt":"The visual classification performance of vision-language models such as CLIP has been shown to benefit from additional semantic knowledge from large language models (LLMs) such as GPT-3. In particular, averaging over LLM-generated class descriptors, e.g. \"waffle, which has a round shape\", can notably improve generalization performance. In this work, we critically study this behavior and propose WaffleCLIP, a framework for zero-shot visual classification which simply replaces LLM-generated descriptors with random character and word descriptors. Without querying external models, we achieve compa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.07282","kind":"arxiv","version":2},"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/2306.07282/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":"2306.07282","created_at":"2026-07-05T06:42:06.334296+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.07282v2","created_at":"2026-07-05T06:42:06.334296+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.07282","created_at":"2026-07-05T06:42:06.334296+00:00"},{"alias_kind":"pith_short_12","alias_value":"RRDKF72LEG4V","created_at":"2026-07-05T06:42:06.334296+00:00"},{"alias_kind":"pith_short_16","alias_value":"RRDKF72LEG4VWBCL","created_at":"2026-07-05T06:42:06.334296+00:00"},{"alias_kind":"pith_short_8","alias_value":"RRDKF72L","created_at":"2026-07-05T06:42:06.334296+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.21786","citing_title":"From Codebooks to VLMs: Evaluating Automated Visual Discourse Analysis for Climate Change on Social Media","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RRDKF72LEG4VWBCL373AKUHUER","json":"https://pith.science/pith/RRDKF72LEG4VWBCL373AKUHUER.json","graph_json":"https://pith.science/api/pith-number/RRDKF72LEG4VWBCL373AKUHUER/graph.json","events_json":"https://pith.science/api/pith-number/RRDKF72LEG4VWBCL373AKUHUER/events.json","paper":"https://pith.science/paper/RRDKF72L"},"agent_actions":{"view_html":"https://pith.science/pith/RRDKF72LEG4VWBCL373AKUHUER","download_json":"https://pith.science/pith/RRDKF72LEG4VWBCL373AKUHUER.json","view_paper":"https://pith.science/paper/RRDKF72L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.07282&json=true","fetch_graph":"https://pith.science/api/pith-number/RRDKF72LEG4VWBCL373AKUHUER/graph.json","fetch_events":"https://pith.science/api/pith-number/RRDKF72LEG4VWBCL373AKUHUER/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RRDKF72LEG4VWBCL373AKUHUER/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RRDKF72LEG4VWBCL373AKUHUER/action/storage_attestation","attest_author":"https://pith.science/pith/RRDKF72LEG4VWBCL373AKUHUER/action/author_attestation","sign_citation":"https://pith.science/pith/RRDKF72LEG4VWBCL373AKUHUER/action/citation_signature","submit_replication":"https://pith.science/pith/RRDKF72LEG4VWBCL373AKUHUER/action/replication_record"}},"created_at":"2026-07-05T06:42:06.334296+00:00","updated_at":"2026-07-05T06:42:06.334296+00:00"}