{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:XPGBUCRZGHD7IZAS7TOYEAHAKL","short_pith_number":"pith:XPGBUCRZ","canonical_record":{"source":{"id":"2605.14689","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T11:03:29Z","cross_cats_sorted":[],"title_canon_sha256":"ff70cd461c2f13b01ddc06b9e8a63fd1281a63f492824590cae2a9a98b35bff5","abstract_canon_sha256":"1a826387b61bb2c300d699f35f0e33166427b0f74f59e4093661feed9f13bef1"},"schema_version":"1.0"},"canonical_sha256":"bbcc1a0a3931c7f46412fcdd8200e052fe3bb1d70d541a17fecfc88797d2412a","source":{"kind":"arxiv","id":"2605.14689","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14689","created_at":"2026-05-17T23:38:59Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14689v1","created_at":"2026-05-17T23:38:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14689","created_at":"2026-05-17T23:38:59Z"},{"alias_kind":"pith_short_12","alias_value":"XPGBUCRZGHD7","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"XPGBUCRZGHD7IZAS","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"XPGBUCRZ","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:XPGBUCRZGHD7IZAS7TOYEAHAKL","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14689","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T11:03:29Z","cross_cats_sorted":[],"title_canon_sha256":"ff70cd461c2f13b01ddc06b9e8a63fd1281a63f492824590cae2a9a98b35bff5","abstract_canon_sha256":"1a826387b61bb2c300d699f35f0e33166427b0f74f59e4093661feed9f13bef1"},"schema_version":"1.0"},"canonical_sha256":"bbcc1a0a3931c7f46412fcdd8200e052fe3bb1d70d541a17fecfc88797d2412a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:59.452170Z","signature_b64":"9ZHyuHii3bnBeNGnWULuk/PvszYsNXYPKnUY6HlaofxfjRmjmF+YvlUX+EcoLtfSo5Gqb7qcJr9uvPogoCAzBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bbcc1a0a3931c7f46412fcdd8200e052fe3bb1d70d541a17fecfc88797d2412a","last_reissued_at":"2026-05-17T23:38:59.451454Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:59.451454Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14689","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RFbnUO7oAckSUtltTMXrORHfFNxx9+Q9CTDTmVocCsaR7KV14mINQFTJm7iUjdUlciOFhfNr/yjb9L8zzZgnBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T21:57:10.765621Z"},"content_sha256":"416b2915797788bb41ae985a5a68a3648b1cf82950ac51767b3e293108165506","schema_version":"1.0","event_id":"sha256:416b2915797788bb41ae985a5a68a3648b1cf82950ac51767b3e293108165506"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:XPGBUCRZGHD7IZAS7TOYEAHAKL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Are Candidate Models Really Needed for Active Learning?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Randomly initialized CNNs and transformers can guide active learning sample selection without any initial candidate models.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Harshini Mridula Mohan, Maanya Manjunath, Nitin Cheekatla, S.H. Shabbeer Basha, Vipul Arya","submitted_at":"2026-05-14T11:03:29Z","abstract_excerpt":"Deep learning has profoundly impacted domains such as computer vision and natural language processing by uncovering complex patterns in vast datasets. However, the reliance on extensive labeled data poses significant challenges, including resource constraints and annotation errors, particularly in training Convolutional Neural Networks (CNNs) and transformers due to a larger number of parameters. Active learning offers a promising solution to reduce labeling burdens by strategically selecting the most informative samples for annotation. However, the current active learning frameworks are time-"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"This study investigates the feasibility of using CNNs and transformers with randomly initialized weights, eliminating the need for initial candidate models while achieving results comparable to active learning frameworks that depend on such candidate models.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That confidence scores produced by randomly initialized CNNs and transformers are sufficiently informative to guide effective sample selection in active learning.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Active learning with randomly initialized models achieves comparable results to traditional candidate-model methods, with low-confidence sampling proving most effective.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Randomly initialized CNNs and transformers can guide active learning sample selection without any initial candidate models.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"c8c0c61549e1d59171dd4704751000cc771963d1310e8cd5efa0341b6a6aa9ba"},"source":{"id":"2605.14689","kind":"arxiv","version":1},"verdict":{"id":"5f715792-7a12-47a4-89c9-a0000bb0ee0e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T05:15:53.247742Z","strongest_claim":"This study investigates the feasibility of using CNNs and transformers with randomly initialized weights, eliminating the need for initial candidate models while achieving results comparable to active learning frameworks that depend on such candidate models.","one_line_summary":"Active learning with randomly initialized models achieves comparable results to traditional candidate-model methods, with low-confidence sampling proving most effective.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That confidence scores produced by randomly initialized CNNs and transformers are sufficiently informative to guide effective sample selection in active learning.","pith_extraction_headline":"Randomly initialized CNNs and transformers can guide active learning sample selection without any initial candidate models."},"references":{"count":180,"sample":[{"doi":"","year":2020,"title":"Journal of Visual Communication and Image Representation , pages=","work_id":"e339097c-a6d0-4d50-ab6f-cb690145ec60","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Dual Swap Disentangling , author=. NeurIPS , year=","work_id":"f1fef762-1b5c-4d50-bb5e-06e3d054a404","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"International Conference on Learning Representations (ICLR) , year=","work_id":"077faf41-b314-486f-b09f-c44d07a0b05a","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"European Conference on Computer Vision (ECCV) , year=","work_id":"8e64195c-772a-41a7-bdd0-5df8f517f486","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"arXiv preprint arXiv:1712.00960 , year=","work_id":"bc6e5180-c013-4ecf-815b-8d934cfc7a1d","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":180,"snapshot_sha256":"c4a15ab1e88ded22be864e43a542482843982e46e6f8474eb00d777c099e7cf9","internal_anchors":16},"formal_canon":{"evidence_count":2,"snapshot_sha256":"534aabf78ed9806f1c8d7bbaf258410700988ca19be69ec9d631199fae2534ed"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"5f715792-7a12-47a4-89c9-a0000bb0ee0e"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"W50omcX2yfWmwNE8M5kNKYausl29YVsPhTSaJFFFeNpG/zM35IoaAzFTUJ8h5mGyj8AGqS+U1jeV8BQweggNDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T21:57:10.766880Z"},"content_sha256":"6bd5d34b1b416a63761b523f5efa842f3c16312a2451d27b25ee91adf8603bde","schema_version":"1.0","event_id":"sha256:6bd5d34b1b416a63761b523f5efa842f3c16312a2451d27b25ee91adf8603bde"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XPGBUCRZGHD7IZAS7TOYEAHAKL/bundle.json","state_url":"https://pith.science/pith/XPGBUCRZGHD7IZAS7TOYEAHAKL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XPGBUCRZGHD7IZAS7TOYEAHAKL/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-31T21:57:10Z","links":{"resolver":"https://pith.science/pith/XPGBUCRZGHD7IZAS7TOYEAHAKL","bundle":"https://pith.science/pith/XPGBUCRZGHD7IZAS7TOYEAHAKL/bundle.json","state":"https://pith.science/pith/XPGBUCRZGHD7IZAS7TOYEAHAKL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XPGBUCRZGHD7IZAS7TOYEAHAKL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:XPGBUCRZGHD7IZAS7TOYEAHAKL","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"1a826387b61bb2c300d699f35f0e33166427b0f74f59e4093661feed9f13bef1","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T11:03:29Z","title_canon_sha256":"ff70cd461c2f13b01ddc06b9e8a63fd1281a63f492824590cae2a9a98b35bff5"},"schema_version":"1.0","source":{"id":"2605.14689","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14689","created_at":"2026-05-17T23:38:59Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14689v1","created_at":"2026-05-17T23:38:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14689","created_at":"2026-05-17T23:38:59Z"},{"alias_kind":"pith_short_12","alias_value":"XPGBUCRZGHD7","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"XPGBUCRZGHD7IZAS","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"XPGBUCRZ","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:6bd5d34b1b416a63761b523f5efa842f3c16312a2451d27b25ee91adf8603bde","target":"graph","created_at":"2026-05-17T23:38:59Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"This study investigates the feasibility of using CNNs and transformers with randomly initialized weights, eliminating the need for initial candidate models while achieving results comparable to active learning frameworks that depend on such candidate models."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That confidence scores produced by randomly initialized CNNs and transformers are sufficiently informative to guide effective sample selection in active learning."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Active learning with randomly initialized models achieves comparable results to traditional candidate-model methods, with low-confidence sampling proving most effective."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Randomly initialized CNNs and transformers can guide active learning sample selection without any initial candidate models."}],"snapshot_sha256":"c8c0c61549e1d59171dd4704751000cc771963d1310e8cd5efa0341b6a6aa9ba"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"534aabf78ed9806f1c8d7bbaf258410700988ca19be69ec9d631199fae2534ed"},"paper":{"abstract_excerpt":"Deep learning has profoundly impacted domains such as computer vision and natural language processing by uncovering complex patterns in vast datasets. However, the reliance on extensive labeled data poses significant challenges, including resource constraints and annotation errors, particularly in training Convolutional Neural Networks (CNNs) and transformers due to a larger number of parameters. Active learning offers a promising solution to reduce labeling burdens by strategically selecting the most informative samples for annotation. However, the current active learning frameworks are time-","authors_text":"Harshini Mridula Mohan, Maanya Manjunath, Nitin Cheekatla, S.H. Shabbeer Basha, Vipul Arya","cross_cats":[],"headline":"Randomly initialized CNNs and transformers can guide active learning sample selection without any initial candidate models.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T11:03:29Z","title":"Are Candidate Models Really Needed for Active Learning?"},"references":{"count":180,"internal_anchors":16,"resolved_work":180,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Journal of Visual Communication and Image Representation , pages=","work_id":"e339097c-a6d0-4d50-ab6f-cb690145ec60","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Dual Swap Disentangling , author=. NeurIPS , year=","work_id":"f1fef762-1b5c-4d50-bb5e-06e3d054a404","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"International Conference on Learning Representations (ICLR) , year=","work_id":"077faf41-b314-486f-b09f-c44d07a0b05a","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"European Conference on Computer Vision (ECCV) , year=","work_id":"8e64195c-772a-41a7-bdd0-5df8f517f486","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"arXiv preprint arXiv:1712.00960 , year=","work_id":"bc6e5180-c013-4ecf-815b-8d934cfc7a1d","year":null}],"snapshot_sha256":"c4a15ab1e88ded22be864e43a542482843982e46e6f8474eb00d777c099e7cf9"},"source":{"id":"2605.14689","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T05:15:53.247742Z","id":"5f715792-7a12-47a4-89c9-a0000bb0ee0e","model_set":{"reader":"grok-4.3"},"one_line_summary":"Active learning with randomly initialized models achieves comparable results to traditional candidate-model methods, with low-confidence sampling proving most effective.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Randomly initialized CNNs and transformers can guide active learning sample selection without any initial candidate models.","strongest_claim":"This study investigates the feasibility of using CNNs and transformers with randomly initialized weights, eliminating the need for initial candidate models while achieving results comparable to active learning frameworks that depend on such candidate models.","weakest_assumption":"That confidence scores produced by randomly initialized CNNs and transformers are sufficiently informative to guide effective sample selection in active learning."}},"verdict_id":"5f715792-7a12-47a4-89c9-a0000bb0ee0e"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:416b2915797788bb41ae985a5a68a3648b1cf82950ac51767b3e293108165506","target":"record","created_at":"2026-05-17T23:38:59Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"1a826387b61bb2c300d699f35f0e33166427b0f74f59e4093661feed9f13bef1","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T11:03:29Z","title_canon_sha256":"ff70cd461c2f13b01ddc06b9e8a63fd1281a63f492824590cae2a9a98b35bff5"},"schema_version":"1.0","source":{"id":"2605.14689","kind":"arxiv","version":1}},"canonical_sha256":"bbcc1a0a3931c7f46412fcdd8200e052fe3bb1d70d541a17fecfc88797d2412a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bbcc1a0a3931c7f46412fcdd8200e052fe3bb1d70d541a17fecfc88797d2412a","first_computed_at":"2026-05-17T23:38:59.451454Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:59.451454Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9ZHyuHii3bnBeNGnWULuk/PvszYsNXYPKnUY6HlaofxfjRmjmF+YvlUX+EcoLtfSo5Gqb7qcJr9uvPogoCAzBg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:59.452170Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14689","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:416b2915797788bb41ae985a5a68a3648b1cf82950ac51767b3e293108165506","sha256:6bd5d34b1b416a63761b523f5efa842f3c16312a2451d27b25ee91adf8603bde"],"state_sha256":"35dc08d0f557a23be756a8969c528460af3cc3d9ebe045c35b524a442deaf014"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IAFbklvsPhMiCtj7j++rxr46p3/HrvpdxWPEQXEjxwcD51D8qiWj5242DwSkhWz/3p9H6Y237X+QD4zMdqhgCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T21:57:10.771588Z","bundle_sha256":"4ad37f153e973a1998c2cf52876b64fc07922069887848e8273bc5f0a5db8935"}}