{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:L4GSSI3YBPYUSV62JMJAVIXT76","short_pith_number":"pith:L4GSSI3Y","schema_version":"1.0","canonical_sha256":"5f0d2923780bf14957da4b120aa2f3ffb371e13a6a6b1fdaeaf8697476836ce5","source":{"kind":"arxiv","id":"2607.22771","version":1},"attestation_state":"computed","paper":{"title":"Cheap Probes Predict Expensive Training in 3D-CT Vision--Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Renjie Liang","submitted_at":"2026-07-24T04:28:33Z","abstract_excerpt":"Picking the frozen image encoder for a 3D~CT vision--language model (VLM), together with the token-compression scheme on top of it, is a search over many candidates. There are several encoders, several ways to compress their tokens, and several token budgets, and the combinations grow fast. Comparing them the usual way means fine-tuning a large language model (LLM) on each combination, and running the whole sweep this way needs far more compute than most groups can spend. We ask whether a cheap probe on the encoder's cached embeddings can stand in for that comparison. We build an image-grounde"},"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":"2607.22771","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-24T04:28:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f61cb9232bdb2389564999e0c17131a4316cc45c2ab919c2d9d804904ddd95e7","abstract_canon_sha256":"6547c4d972201cc2ef933868226fa4a91dc9981a7741f19929d5b096df44435f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T00:21:53.546708Z","signature_b64":"vFKOjaJcJfnj2fTE+tQckg5eajcE56V/LU2UQp7BTkmpu82gnPd/f6m5PFaXmL3ne8C1feUtwpMPLGyrDyjCDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f0d2923780bf14957da4b120aa2f3ffb371e13a6a6b1fdaeaf8697476836ce5","last_reissued_at":"2026-07-28T00:21:53.545742Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T00:21:53.545742Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cheap Probes Predict Expensive Training in 3D-CT Vision--Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Renjie Liang","submitted_at":"2026-07-24T04:28:33Z","abstract_excerpt":"Picking the frozen image encoder for a 3D~CT vision--language model (VLM), together with the token-compression scheme on top of it, is a search over many candidates. There are several encoders, several ways to compress their tokens, and several token budgets, and the combinations grow fast. Comparing them the usual way means fine-tuning a large language model (LLM) on each combination, and running the whole sweep this way needs far more compute than most groups can spend. We ask whether a cheap probe on the encoder's cached embeddings can stand in for that comparison. We build an image-grounde"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.22771","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/2607.22771/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":"2607.22771","created_at":"2026-07-28T00:21:53.546201+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.22771v1","created_at":"2026-07-28T00:21:53.546201+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.22771","created_at":"2026-07-28T00:21:53.546201+00:00"},{"alias_kind":"pith_short_12","alias_value":"L4GSSI3YBPYU","created_at":"2026-07-28T00:21:53.546201+00:00"},{"alias_kind":"pith_short_16","alias_value":"L4GSSI3YBPYUSV62","created_at":"2026-07-28T00:21:53.546201+00:00"},{"alias_kind":"pith_short_8","alias_value":"L4GSSI3Y","created_at":"2026-07-28T00:21:53.546201+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L4GSSI3YBPYUSV62JMJAVIXT76","json":"https://pith.science/pith/L4GSSI3YBPYUSV62JMJAVIXT76.json","graph_json":"https://pith.science/api/pith-number/L4GSSI3YBPYUSV62JMJAVIXT76/graph.json","events_json":"https://pith.science/api/pith-number/L4GSSI3YBPYUSV62JMJAVIXT76/events.json","paper":"https://pith.science/paper/L4GSSI3Y"},"agent_actions":{"view_html":"https://pith.science/pith/L4GSSI3YBPYUSV62JMJAVIXT76","download_json":"https://pith.science/pith/L4GSSI3YBPYUSV62JMJAVIXT76.json","view_paper":"https://pith.science/paper/L4GSSI3Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.22771&json=true","fetch_graph":"https://pith.science/api/pith-number/L4GSSI3YBPYUSV62JMJAVIXT76/graph.json","fetch_events":"https://pith.science/api/pith-number/L4GSSI3YBPYUSV62JMJAVIXT76/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L4GSSI3YBPYUSV62JMJAVIXT76/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L4GSSI3YBPYUSV62JMJAVIXT76/action/storage_attestation","attest_author":"https://pith.science/pith/L4GSSI3YBPYUSV62JMJAVIXT76/action/author_attestation","sign_citation":"https://pith.science/pith/L4GSSI3YBPYUSV62JMJAVIXT76/action/citation_signature","submit_replication":"https://pith.science/pith/L4GSSI3YBPYUSV62JMJAVIXT76/action/replication_record"}},"created_at":"2026-07-28T00:21:53.546201+00:00","updated_at":"2026-07-28T00:21:53.546201+00:00"}