{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:4TNIZV4RA4BOOETU6S5SOZWRP3","short_pith_number":"pith:4TNIZV4R","schema_version":"1.0","canonical_sha256":"e4da8cd7910702e71274f4bb2766d17ee9d7723ce73614def10db0dc7b6df9c8","source":{"kind":"arxiv","id":"2403.17343","version":3},"attestation_state":"computed","paper":{"title":"Residual-based Language Models are Free Boosters for Biomedical Imaging","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Jing Wu, Naira Hovakimyan, Suiyao Chen, Yucheng Zhou, Zhixin Lai","submitted_at":"2024-03-26T03:05:20Z","abstract_excerpt":"In this study, we uncover the unexpected efficacy of residual-based large language models (LLMs) as part of encoders for biomedical imaging tasks, a domain traditionally devoid of language or textual data. The approach diverges from established methodologies by utilizing a frozen transformer block, extracted from pre-trained LLMs, as an innovative encoder layer for the direct processing of visual tokens. This strategy represents a significant departure from the standard multi-modal vision-language frameworks, which typically hinge on language-driven prompts and inputs. We found that these LLMs"},"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":"2403.17343","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-26T03:05:20Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"7f6715c6be34930725e755ebb81cbb13df6d258400df910d6ac641c5a6ebd562","abstract_canon_sha256":"4cba36541370ba3b8d35afc6d14d87206d3de7cb74893d3c2cdcf23b98b75bed"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:02:06.479677Z","signature_b64":"7wUUjz4onRMsGqTbOb8vpFP4axIO9DQ+J4MJxHNfZyngldNnUJGVZuEERRz0MAFOFE4PpjftlUBMz5HnhPEnCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4da8cd7910702e71274f4bb2766d17ee9d7723ce73614def10db0dc7b6df9c8","last_reissued_at":"2026-07-05T08:02:06.479000Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:02:06.479000Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Residual-based Language Models are Free Boosters for Biomedical Imaging","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Jing Wu, Naira Hovakimyan, Suiyao Chen, Yucheng Zhou, Zhixin Lai","submitted_at":"2024-03-26T03:05:20Z","abstract_excerpt":"In this study, we uncover the unexpected efficacy of residual-based large language models (LLMs) as part of encoders for biomedical imaging tasks, a domain traditionally devoid of language or textual data. The approach diverges from established methodologies by utilizing a frozen transformer block, extracted from pre-trained LLMs, as an innovative encoder layer for the direct processing of visual tokens. This strategy represents a significant departure from the standard multi-modal vision-language frameworks, which typically hinge on language-driven prompts and inputs. We found that these LLMs"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.17343","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/2403.17343/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":"2403.17343","created_at":"2026-07-05T08:02:06.479080+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.17343v3","created_at":"2026-07-05T08:02:06.479080+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.17343","created_at":"2026-07-05T08:02:06.479080+00:00"},{"alias_kind":"pith_short_12","alias_value":"4TNIZV4RA4BO","created_at":"2026-07-05T08:02:06.479080+00:00"},{"alias_kind":"pith_short_16","alias_value":"4TNIZV4RA4BOOETU","created_at":"2026-07-05T08:02:06.479080+00:00"},{"alias_kind":"pith_short_8","alias_value":"4TNIZV4R","created_at":"2026-07-05T08:02:06.479080+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/4TNIZV4RA4BOOETU6S5SOZWRP3","json":"https://pith.science/pith/4TNIZV4RA4BOOETU6S5SOZWRP3.json","graph_json":"https://pith.science/api/pith-number/4TNIZV4RA4BOOETU6S5SOZWRP3/graph.json","events_json":"https://pith.science/api/pith-number/4TNIZV4RA4BOOETU6S5SOZWRP3/events.json","paper":"https://pith.science/paper/4TNIZV4R"},"agent_actions":{"view_html":"https://pith.science/pith/4TNIZV4RA4BOOETU6S5SOZWRP3","download_json":"https://pith.science/pith/4TNIZV4RA4BOOETU6S5SOZWRP3.json","view_paper":"https://pith.science/paper/4TNIZV4R","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.17343&json=true","fetch_graph":"https://pith.science/api/pith-number/4TNIZV4RA4BOOETU6S5SOZWRP3/graph.json","fetch_events":"https://pith.science/api/pith-number/4TNIZV4RA4BOOETU6S5SOZWRP3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4TNIZV4RA4BOOETU6S5SOZWRP3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4TNIZV4RA4BOOETU6S5SOZWRP3/action/storage_attestation","attest_author":"https://pith.science/pith/4TNIZV4RA4BOOETU6S5SOZWRP3/action/author_attestation","sign_citation":"https://pith.science/pith/4TNIZV4RA4BOOETU6S5SOZWRP3/action/citation_signature","submit_replication":"https://pith.science/pith/4TNIZV4RA4BOOETU6S5SOZWRP3/action/replication_record"}},"created_at":"2026-07-05T08:02:06.479080+00:00","updated_at":"2026-07-05T08:02:06.479080+00:00"}