{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:75RCP27GN26O4JFZC32T3ILR5M","short_pith_number":"pith:75RCP27G","schema_version":"1.0","canonical_sha256":"ff6227ebe66ebcee24b916f53da171eb12febd10af85d2ba537bb9c7a40d6fa0","source":{"kind":"arxiv","id":"2404.00701","version":1},"attestation_state":"computed","paper":{"title":"Training-Free Semantic Segmentation via LLM-Supervision","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cees G.M. Snoek, Gaowen Liu, Ramana Kompella, Wenfang Sun, Yingjun Du","submitted_at":"2024-03-31T14:37:25Z","abstract_excerpt":"Recent advancements in open vocabulary models, like CLIP, have notably advanced zero-shot classification and segmentation by utilizing natural language for class-specific embeddings. However, most research has focused on improving model accuracy through prompt engineering, prompt learning, or fine-tuning with limited labeled data, thereby overlooking the importance of refining the class descriptors. This paper introduces a new approach to text-supervised semantic segmentation using supervision by a large language model (LLM) that does not require extra training. Our method starts from an LLM, "},"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":"2404.00701","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-31T14:37:25Z","cross_cats_sorted":[],"title_canon_sha256":"48f32bc50e7586bd85843ed2788979d17f0939dd614a81066220a5fe6aff4a30","abstract_canon_sha256":"97ed55f99eda0752f2acb93befdd8d144b5181dd85a19fff62a2b5bed3284a72"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:02:55.553216Z","signature_b64":"zoDBm3KzbxGg7MdjL3hg0rXSF84T83wi1JeQ+UsXlmG6QLG5ot4LJXIu0mjQW1q+reGSib0lw/cxOzoMt/4YDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff6227ebe66ebcee24b916f53da171eb12febd10af85d2ba537bb9c7a40d6fa0","last_reissued_at":"2026-07-05T08:02:55.552754Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:02:55.552754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Training-Free Semantic Segmentation via LLM-Supervision","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cees G.M. Snoek, Gaowen Liu, Ramana Kompella, Wenfang Sun, Yingjun Du","submitted_at":"2024-03-31T14:37:25Z","abstract_excerpt":"Recent advancements in open vocabulary models, like CLIP, have notably advanced zero-shot classification and segmentation by utilizing natural language for class-specific embeddings. However, most research has focused on improving model accuracy through prompt engineering, prompt learning, or fine-tuning with limited labeled data, thereby overlooking the importance of refining the class descriptors. This paper introduces a new approach to text-supervised semantic segmentation using supervision by a large language model (LLM) that does not require extra training. Our method starts from an LLM, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.00701","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/2404.00701/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":"2404.00701","created_at":"2026-07-05T08:02:55.552807+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.00701v1","created_at":"2026-07-05T08:02:55.552807+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.00701","created_at":"2026-07-05T08:02:55.552807+00:00"},{"alias_kind":"pith_short_12","alias_value":"75RCP27GN26O","created_at":"2026-07-05T08:02:55.552807+00:00"},{"alias_kind":"pith_short_16","alias_value":"75RCP27GN26O4JFZ","created_at":"2026-07-05T08:02:55.552807+00:00"},{"alias_kind":"pith_short_8","alias_value":"75RCP27G","created_at":"2026-07-05T08:02:55.552807+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.12325","citing_title":"VIP: Visual-guided Prompt Evolution for Efficient Dense Vision-Language Inference","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12325","citing_title":"VIP: Visual-guided Prompt Evolution for Efficient Dense Vision-Language Inference","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/75RCP27GN26O4JFZC32T3ILR5M","json":"https://pith.science/pith/75RCP27GN26O4JFZC32T3ILR5M.json","graph_json":"https://pith.science/api/pith-number/75RCP27GN26O4JFZC32T3ILR5M/graph.json","events_json":"https://pith.science/api/pith-number/75RCP27GN26O4JFZC32T3ILR5M/events.json","paper":"https://pith.science/paper/75RCP27G"},"agent_actions":{"view_html":"https://pith.science/pith/75RCP27GN26O4JFZC32T3ILR5M","download_json":"https://pith.science/pith/75RCP27GN26O4JFZC32T3ILR5M.json","view_paper":"https://pith.science/paper/75RCP27G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.00701&json=true","fetch_graph":"https://pith.science/api/pith-number/75RCP27GN26O4JFZC32T3ILR5M/graph.json","fetch_events":"https://pith.science/api/pith-number/75RCP27GN26O4JFZC32T3ILR5M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/75RCP27GN26O4JFZC32T3ILR5M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/75RCP27GN26O4JFZC32T3ILR5M/action/storage_attestation","attest_author":"https://pith.science/pith/75RCP27GN26O4JFZC32T3ILR5M/action/author_attestation","sign_citation":"https://pith.science/pith/75RCP27GN26O4JFZC32T3ILR5M/action/citation_signature","submit_replication":"https://pith.science/pith/75RCP27GN26O4JFZC32T3ILR5M/action/replication_record"}},"created_at":"2026-07-05T08:02:55.552807+00:00","updated_at":"2026-07-05T08:02:55.552807+00:00"}