{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PHEXP55MURKCWIAMDU6YIYR5VA","short_pith_number":"pith:PHEXP55M","schema_version":"1.0","canonical_sha256":"79c977f7aca4542b200c1d3d84623da829219f0677b1560c0a1d058f7e132411","source":{"kind":"arxiv","id":"2506.15681","version":4},"attestation_state":"computed","paper":{"title":"GenRecal: Generation after Recalibration from Large to Small Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Byung-Kwan Lee, Ryo Hachiuma, Yong Man Ro, Yu-Chiang Frank Wang, Yueh-Hua Wu","submitted_at":"2025-06-18T17:59:49Z","abstract_excerpt":"Recent advancements in vision-language models (VLMs) have leveraged large language models (LLMs) to achieve performance on par with closed-source systems like GPT-4V. However, deploying these models in real-world scenarios, particularly on resource-constrained devices, remains challenging due to their substantial computational demands. This has spurred interest in distilling knowledge from large VLMs into smaller, more efficient counterparts. A key challenge arises here from the diversity of VLM architectures, which are built on different LLMs and employ varying token types-differing in vocabu"},"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":"2506.15681","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-18T17:59:49Z","cross_cats_sorted":[],"title_canon_sha256":"ab5505392649c3d573ccb04c78d7403fbc7109b5580d6a846506d0fa11a8a97a","abstract_canon_sha256":"5f89ad902559f7f87fc1b2aca7c7ff2795c17479cd88a804f80fffc439d93824"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-26T01:15:44.614336Z","signature_b64":"7FoE7VsHrdLKc829SpaaRi9/PHLE4SmNsPtLFJfrx0kpHe4T2N1L4YBMkhsHu7cE4JRp9QnNCslSNrSvP7bBCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"79c977f7aca4542b200c1d3d84623da829219f0677b1560c0a1d058f7e132411","last_reissued_at":"2026-06-26T01:15:44.613792Z","signature_status":"signed_v1","first_computed_at":"2026-06-26T01:15:44.613792Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GenRecal: Generation after Recalibration from Large to Small Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Byung-Kwan Lee, Ryo Hachiuma, Yong Man Ro, Yu-Chiang Frank Wang, Yueh-Hua Wu","submitted_at":"2025-06-18T17:59:49Z","abstract_excerpt":"Recent advancements in vision-language models (VLMs) have leveraged large language models (LLMs) to achieve performance on par with closed-source systems like GPT-4V. However, deploying these models in real-world scenarios, particularly on resource-constrained devices, remains challenging due to their substantial computational demands. This has spurred interest in distilling knowledge from large VLMs into smaller, more efficient counterparts. A key challenge arises here from the diversity of VLM architectures, which are built on different LLMs and employ varying token types-differing in vocabu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.15681","kind":"arxiv","version":4},"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/2506.15681/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":"2506.15681","created_at":"2026-06-26T01:15:44.613849+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.15681v4","created_at":"2026-06-26T01:15:44.613849+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.15681","created_at":"2026-06-26T01:15:44.613849+00:00"},{"alias_kind":"pith_short_12","alias_value":"PHEXP55MURKC","created_at":"2026-06-26T01:15:44.613849+00:00"},{"alias_kind":"pith_short_16","alias_value":"PHEXP55MURKCWIAM","created_at":"2026-06-26T01:15:44.613849+00:00"},{"alias_kind":"pith_short_8","alias_value":"PHEXP55M","created_at":"2026-06-26T01:15:44.613849+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2606.29082","citing_title":"Evolution Fine-Tuning: Learning to Discover Across 371 Optimization Tasks","ref_index":47,"is_internal_anchor":true},{"citing_arxiv_id":"2605.28774","citing_title":"Agent Explorative Policy Optimization for Multimodal Agentic Reasoning","ref_index":81,"is_internal_anchor":true},{"citing_arxiv_id":"2604.12358","citing_title":"Why and When Visual Token Pruning Fails? A Study on Relevant Visual Information Shift in MLLMs Decoding","ref_index":21,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PHEXP55MURKCWIAMDU6YIYR5VA","json":"https://pith.science/pith/PHEXP55MURKCWIAMDU6YIYR5VA.json","graph_json":"https://pith.science/api/pith-number/PHEXP55MURKCWIAMDU6YIYR5VA/graph.json","events_json":"https://pith.science/api/pith-number/PHEXP55MURKCWIAMDU6YIYR5VA/events.json","paper":"https://pith.science/paper/PHEXP55M"},"agent_actions":{"view_html":"https://pith.science/pith/PHEXP55MURKCWIAMDU6YIYR5VA","download_json":"https://pith.science/pith/PHEXP55MURKCWIAMDU6YIYR5VA.json","view_paper":"https://pith.science/paper/PHEXP55M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.15681&json=true","fetch_graph":"https://pith.science/api/pith-number/PHEXP55MURKCWIAMDU6YIYR5VA/graph.json","fetch_events":"https://pith.science/api/pith-number/PHEXP55MURKCWIAMDU6YIYR5VA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PHEXP55MURKCWIAMDU6YIYR5VA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PHEXP55MURKCWIAMDU6YIYR5VA/action/storage_attestation","attest_author":"https://pith.science/pith/PHEXP55MURKCWIAMDU6YIYR5VA/action/author_attestation","sign_citation":"https://pith.science/pith/PHEXP55MURKCWIAMDU6YIYR5VA/action/citation_signature","submit_replication":"https://pith.science/pith/PHEXP55MURKCWIAMDU6YIYR5VA/action/replication_record"}},"created_at":"2026-06-26T01:15:44.613849+00:00","updated_at":"2026-06-26T01:15:44.613849+00:00"}