{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:MXLHYVBT3IDAF4TBEU4LACFOVP","short_pith_number":"pith:MXLHYVBT","schema_version":"1.0","canonical_sha256":"65d67c5433da0602f2612538b008aeabe30e1d5f99b7aa4ff77f5aa1b289f6a4","source":{"kind":"arxiv","id":"2404.01650","version":2},"attestation_state":"computed","paper":{"title":"Test-Time Model Adaptation with Only Forward Passes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chunyan Miao, Guohao Chen, Peilin Zhao, Pengcheng Wu, Shuaicheng Niu","submitted_at":"2024-04-02T05:34:33Z","abstract_excerpt":"Test-time adaptation has proven effective in adapting a given trained model to unseen test samples with potential distribution shifts. However, in real-world scenarios, models are usually deployed on resource-limited devices, e.g., FPGAs, and are often quantized and hard-coded with non-modifiable parameters for acceleration. In light of this, existing methods are often infeasible since they heavily depend on computation-intensive backpropagation for model updating that may be not supported. To address this, we propose a test-time Forward-Optimization Adaptation (FOA) method. In FOA, we seek to"},"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":"2404.01650","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-02T05:34:33Z","cross_cats_sorted":[],"title_canon_sha256":"ed457645c99fe06c296f8e78fc17aadc2833eba71e4f3161fb8d4679b742cc8a","abstract_canon_sha256":"2d7041b06813f3b9e195fe88c6fe89a4e6a461aedb5d3ac8980bc16d3d2c7306"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:24:24.545623Z","signature_b64":"gHlmg6yAMgoFBkNWlmMKrS3ezmjCrH91dfO9gAsDIC1hV3Hds2AE2K/EfXqhyLW16LdCQJyJ1n653axKvVbCBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"65d67c5433da0602f2612538b008aeabe30e1d5f99b7aa4ff77f5aa1b289f6a4","last_reissued_at":"2026-07-05T08:24:24.545154Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:24:24.545154Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Test-Time Model Adaptation with Only Forward Passes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chunyan Miao, Guohao Chen, Peilin Zhao, Pengcheng Wu, Shuaicheng Niu","submitted_at":"2024-04-02T05:34:33Z","abstract_excerpt":"Test-time adaptation has proven effective in adapting a given trained model to unseen test samples with potential distribution shifts. However, in real-world scenarios, models are usually deployed on resource-limited devices, e.g., FPGAs, and are often quantized and hard-coded with non-modifiable parameters for acceleration. In light of this, existing methods are often infeasible since they heavily depend on computation-intensive backpropagation for model updating that may be not supported. To address this, we propose a test-time Forward-Optimization Adaptation (FOA) method. In FOA, we seek to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.01650","kind":"arxiv","version":2},"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/2404.01650/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":"2404.01650","created_at":"2026-07-05T08:24:24.545216+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.01650v2","created_at":"2026-07-05T08:24:24.545216+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.01650","created_at":"2026-07-05T08:24:24.545216+00:00"},{"alias_kind":"pith_short_12","alias_value":"MXLHYVBT3IDA","created_at":"2026-07-05T08:24:24.545216+00:00"},{"alias_kind":"pith_short_16","alias_value":"MXLHYVBT3IDAF4TB","created_at":"2026-07-05T08:24:24.545216+00:00"},{"alias_kind":"pith_short_8","alias_value":"MXLHYVBT","created_at":"2026-07-05T08:24:24.545216+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07646","citing_title":"DOME: Learning Transferable Domain Variables from Sparse Supervision for Test-Time Adaptation","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18608","citing_title":"Dance Across Shifts: Forward-Facilitation Continual Test-Time Adaptation through Dynamic Style Bridging","ref_index":45,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MXLHYVBT3IDAF4TBEU4LACFOVP","json":"https://pith.science/pith/MXLHYVBT3IDAF4TBEU4LACFOVP.json","graph_json":"https://pith.science/api/pith-number/MXLHYVBT3IDAF4TBEU4LACFOVP/graph.json","events_json":"https://pith.science/api/pith-number/MXLHYVBT3IDAF4TBEU4LACFOVP/events.json","paper":"https://pith.science/paper/MXLHYVBT"},"agent_actions":{"view_html":"https://pith.science/pith/MXLHYVBT3IDAF4TBEU4LACFOVP","download_json":"https://pith.science/pith/MXLHYVBT3IDAF4TBEU4LACFOVP.json","view_paper":"https://pith.science/paper/MXLHYVBT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.01650&json=true","fetch_graph":"https://pith.science/api/pith-number/MXLHYVBT3IDAF4TBEU4LACFOVP/graph.json","fetch_events":"https://pith.science/api/pith-number/MXLHYVBT3IDAF4TBEU4LACFOVP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MXLHYVBT3IDAF4TBEU4LACFOVP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MXLHYVBT3IDAF4TBEU4LACFOVP/action/storage_attestation","attest_author":"https://pith.science/pith/MXLHYVBT3IDAF4TBEU4LACFOVP/action/author_attestation","sign_citation":"https://pith.science/pith/MXLHYVBT3IDAF4TBEU4LACFOVP/action/citation_signature","submit_replication":"https://pith.science/pith/MXLHYVBT3IDAF4TBEU4LACFOVP/action/replication_record"}},"created_at":"2026-07-05T08:24:24.545216+00:00","updated_at":"2026-07-05T08:24:24.545216+00:00"}