{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:CMSZOGTZMBD4ESXCWMBBKGHZOW","short_pith_number":"pith:CMSZOGTZ","schema_version":"1.0","canonical_sha256":"1325971a796047c24ae2b3021518f975969f19940eff23a65cdfbbd65a1a11ed","source":{"kind":"arxiv","id":"2210.07795","version":1},"attestation_state":"computed","paper":{"title":"EfficientVLM: Fast and Accurate Vision-Language Models via Knowledge Distillation and Modal-adaptive Pruning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Tiannan Wang, Wangchunshu Zhou, Xinsong Zhang, Yan Zeng","submitted_at":"2022-10-14T13:26:41Z","abstract_excerpt":"Pre-trained vision-language models (VLMs) have achieved impressive results in a range of vision-language tasks. However, popular VLMs usually consist of hundreds of millions of parameters which brings challenges for fine-tuning and deployment in real-world applications due to space, memory, and latency constraints. In this work, we introduce a distilling then pruning framework to compress large vision-language models into smaller, faster, and more accurate ones. We first shrink the size of a pre-trained large VLM and apply knowledge distillation in the vision-language pre-training stage to obt"},"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":"2210.07795","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-10-14T13:26:41Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"834a2ba721f638b2d34e6ab0127b73e9f95303b6771d499883e6003f74532198","abstract_canon_sha256":"fd98374e904c2c512737e4048a03b18e4a83c28acd37d3b4761618f94cf948e7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:06:41.066500Z","signature_b64":"Fy1s6hTSIO3XS+EroOxX8OS3Gx2xAEBiRqHitbPkiKsmGgrSejtPps0Qa4UTTl/rrqwi8BvYO3vhG7/AyjPiCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1325971a796047c24ae2b3021518f975969f19940eff23a65cdfbbd65a1a11ed","last_reissued_at":"2026-07-05T05:06:41.066082Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:06:41.066082Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EfficientVLM: Fast and Accurate Vision-Language Models via Knowledge Distillation and Modal-adaptive Pruning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Tiannan Wang, Wangchunshu Zhou, Xinsong Zhang, Yan Zeng","submitted_at":"2022-10-14T13:26:41Z","abstract_excerpt":"Pre-trained vision-language models (VLMs) have achieved impressive results in a range of vision-language tasks. However, popular VLMs usually consist of hundreds of millions of parameters which brings challenges for fine-tuning and deployment in real-world applications due to space, memory, and latency constraints. In this work, we introduce a distilling then pruning framework to compress large vision-language models into smaller, faster, and more accurate ones. We first shrink the size of a pre-trained large VLM and apply knowledge distillation in the vision-language pre-training stage to obt"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.07795","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/2210.07795/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":"2210.07795","created_at":"2026-07-05T05:06:41.066137+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.07795v1","created_at":"2026-07-05T05:06:41.066137+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.07795","created_at":"2026-07-05T05:06:41.066137+00:00"},{"alias_kind":"pith_short_12","alias_value":"CMSZOGTZMBD4","created_at":"2026-07-05T05:06:41.066137+00:00"},{"alias_kind":"pith_short_16","alias_value":"CMSZOGTZMBD4ESXC","created_at":"2026-07-05T05:06:41.066137+00:00"},{"alias_kind":"pith_short_8","alias_value":"CMSZOGTZ","created_at":"2026-07-05T05:06:41.066137+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.18117","citing_title":"Online In-Context Distillation for Low-Resource Vision Language Models","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2311.01378","citing_title":"Vision-Language Foundation Models as Effective Robot Imitators","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18452","citing_title":"ESsEN: Training Compact Discriminative Vision-Language Transformers in a Low-Resource Setting","ref_index":49,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CMSZOGTZMBD4ESXCWMBBKGHZOW","json":"https://pith.science/pith/CMSZOGTZMBD4ESXCWMBBKGHZOW.json","graph_json":"https://pith.science/api/pith-number/CMSZOGTZMBD4ESXCWMBBKGHZOW/graph.json","events_json":"https://pith.science/api/pith-number/CMSZOGTZMBD4ESXCWMBBKGHZOW/events.json","paper":"https://pith.science/paper/CMSZOGTZ"},"agent_actions":{"view_html":"https://pith.science/pith/CMSZOGTZMBD4ESXCWMBBKGHZOW","download_json":"https://pith.science/pith/CMSZOGTZMBD4ESXCWMBBKGHZOW.json","view_paper":"https://pith.science/paper/CMSZOGTZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.07795&json=true","fetch_graph":"https://pith.science/api/pith-number/CMSZOGTZMBD4ESXCWMBBKGHZOW/graph.json","fetch_events":"https://pith.science/api/pith-number/CMSZOGTZMBD4ESXCWMBBKGHZOW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CMSZOGTZMBD4ESXCWMBBKGHZOW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CMSZOGTZMBD4ESXCWMBBKGHZOW/action/storage_attestation","attest_author":"https://pith.science/pith/CMSZOGTZMBD4ESXCWMBBKGHZOW/action/author_attestation","sign_citation":"https://pith.science/pith/CMSZOGTZMBD4ESXCWMBBKGHZOW/action/citation_signature","submit_replication":"https://pith.science/pith/CMSZOGTZMBD4ESXCWMBBKGHZOW/action/replication_record"}},"created_at":"2026-07-05T05:06:41.066137+00:00","updated_at":"2026-07-05T05:06:41.066137+00:00"}