{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZFNVIKQIK36NIXCAT7WNETGEGL","short_pith_number":"pith:ZFNVIKQI","schema_version":"1.0","canonical_sha256":"c95b542a0856fcd45c409fecd24cc432fc9292d98c6be58231b3e926e9ce3209","source":{"kind":"arxiv","id":"2607.02158","version":1},"attestation_state":"computed","paper":{"title":"Efficient PEFT Methods with Adaptive Checkpointing for Vision Models and VLMs on Resource Constrained Consumer-GPUs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Altay Toktassyn, Jurn-Gyu Park","submitted_at":"2026-07-02T13:31:29Z","abstract_excerpt":"Modern pretrained vision models achieve strong accuracy but demand substantial GPU memory for fine-tuning, making edge deployment impractical. This paper compares five parameter-efficient fine-tuning (PEFT) methods (Full FT, LoRA, AdaLoRA, QLoRA, BitFit) on Transformers- (ViT-Small, TinyViT) and Mamba-based vision backbones (Vim-Small, MambaVision-T) under an on-device VRAM budget (e.g., 2 GB), together with three gradient-checkpointing strategies (none, static, and a proposed memory-budget-aware adaptive algorithm); and we evaluate three families of foundation-model baselines: zero-shot contr"},"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":"2607.02158","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-02T13:31:29Z","cross_cats_sorted":[],"title_canon_sha256":"606b87558c0ce5aa63db69a4bea94cc154e634b255152155a77c84bb94b91587","abstract_canon_sha256":"478a948f205051e0d12fc48252e0ec3a90d9c083fe33a4e5200725b4a2697331"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T01:17:43.624046Z","signature_b64":"dGGSLmWZ/oGBQ6Xkuf+0wRjfe1yA70ohCUUlddLz6ME4fqJsNIgRnV3dhXFerrp91O9AkDJQTWqoLs/jFE7WAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c95b542a0856fcd45c409fecd24cc432fc9292d98c6be58231b3e926e9ce3209","last_reissued_at":"2026-07-03T01:17:43.623676Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T01:17:43.623676Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient PEFT Methods with Adaptive Checkpointing for Vision Models and VLMs on Resource Constrained Consumer-GPUs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Altay Toktassyn, Jurn-Gyu Park","submitted_at":"2026-07-02T13:31:29Z","abstract_excerpt":"Modern pretrained vision models achieve strong accuracy but demand substantial GPU memory for fine-tuning, making edge deployment impractical. This paper compares five parameter-efficient fine-tuning (PEFT) methods (Full FT, LoRA, AdaLoRA, QLoRA, BitFit) on Transformers- (ViT-Small, TinyViT) and Mamba-based vision backbones (Vim-Small, MambaVision-T) under an on-device VRAM budget (e.g., 2 GB), together with three gradient-checkpointing strategies (none, static, and a proposed memory-budget-aware adaptive algorithm); and we evaluate three families of foundation-model baselines: zero-shot contr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02158","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/2607.02158/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":"2607.02158","created_at":"2026-07-03T01:17:43.623733+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.02158v1","created_at":"2026-07-03T01:17:43.623733+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02158","created_at":"2026-07-03T01:17:43.623733+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZFNVIKQIK36N","created_at":"2026-07-03T01:17:43.623733+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZFNVIKQIK36NIXCA","created_at":"2026-07-03T01:17:43.623733+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZFNVIKQI","created_at":"2026-07-03T01:17:43.623733+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/ZFNVIKQIK36NIXCAT7WNETGEGL","json":"https://pith.science/pith/ZFNVIKQIK36NIXCAT7WNETGEGL.json","graph_json":"https://pith.science/api/pith-number/ZFNVIKQIK36NIXCAT7WNETGEGL/graph.json","events_json":"https://pith.science/api/pith-number/ZFNVIKQIK36NIXCAT7WNETGEGL/events.json","paper":"https://pith.science/paper/ZFNVIKQI"},"agent_actions":{"view_html":"https://pith.science/pith/ZFNVIKQIK36NIXCAT7WNETGEGL","download_json":"https://pith.science/pith/ZFNVIKQIK36NIXCAT7WNETGEGL.json","view_paper":"https://pith.science/paper/ZFNVIKQI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.02158&json=true","fetch_graph":"https://pith.science/api/pith-number/ZFNVIKQIK36NIXCAT7WNETGEGL/graph.json","fetch_events":"https://pith.science/api/pith-number/ZFNVIKQIK36NIXCAT7WNETGEGL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZFNVIKQIK36NIXCAT7WNETGEGL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZFNVIKQIK36NIXCAT7WNETGEGL/action/storage_attestation","attest_author":"https://pith.science/pith/ZFNVIKQIK36NIXCAT7WNETGEGL/action/author_attestation","sign_citation":"https://pith.science/pith/ZFNVIKQIK36NIXCAT7WNETGEGL/action/citation_signature","submit_replication":"https://pith.science/pith/ZFNVIKQIK36NIXCAT7WNETGEGL/action/replication_record"}},"created_at":"2026-07-03T01:17:43.623733+00:00","updated_at":"2026-07-03T01:17:43.623733+00:00"}