{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:MCZAZT5DFT5JOL53SNUBNORC6E","short_pith_number":"pith:MCZAZT5D","schema_version":"1.0","canonical_sha256":"60b20ccfa32cfa972fbb936816ba22f13361ecb2ecefd7ede1c301f2d9829831","source":{"kind":"arxiv","id":"2005.00060","version":2},"attestation_state":"computed","paper":{"title":"Bridging Mode Connectivity in Loss Landscapes and Adversarial Robustness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Karthikeyan Natesan Ramamurthy, Payel Das, Pin-Yu Chen, Pu Zhao, Xue Lin","submitted_at":"2020-04-30T19:12:50Z","abstract_excerpt":"Mode connectivity provides novel geometric insights on analyzing loss landscapes and enables building high-accuracy pathways between well-trained neural networks. In this work, we propose to employ mode connectivity in loss landscapes to study the adversarial robustness of deep neural networks, and provide novel methods for improving this robustness. Our experiments cover various types of adversarial attacks applied to different network architectures and datasets. When network models are tampered with backdoor or error-injection attacks, our results demonstrate that the path connection learned"},"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":"2005.00060","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-04-30T19:12:50Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"ce452a107f4467c69e4ef92093d8766679b7a6ab34e4dbdde25ef4dca5d77320","abstract_canon_sha256":"2c5fb1db931787c6d461a5fd2d897f99d9c7f7958d71fc02e4ef5bcdd75043f1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:15:58.643246Z","signature_b64":"l+lmEiH5AgjFN10ORSzo2iHAhqr1/jP+biwwJ/oSZrAmX/7DSfDV+DaoTmyS61bZy6msNoDfLUYBtJWktRwRCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"60b20ccfa32cfa972fbb936816ba22f13361ecb2ecefd7ede1c301f2d9829831","last_reissued_at":"2026-07-05T01:15:58.642784Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:15:58.642784Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bridging Mode Connectivity in Loss Landscapes and Adversarial Robustness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Karthikeyan Natesan Ramamurthy, Payel Das, Pin-Yu Chen, Pu Zhao, Xue Lin","submitted_at":"2020-04-30T19:12:50Z","abstract_excerpt":"Mode connectivity provides novel geometric insights on analyzing loss landscapes and enables building high-accuracy pathways between well-trained neural networks. In this work, we propose to employ mode connectivity in loss landscapes to study the adversarial robustness of deep neural networks, and provide novel methods for improving this robustness. Our experiments cover various types of adversarial attacks applied to different network architectures and datasets. When network models are tampered with backdoor or error-injection attacks, our results demonstrate that the path connection learned"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.00060","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/2005.00060/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":"2005.00060","created_at":"2026-07-05T01:15:58.642847+00:00"},{"alias_kind":"arxiv_version","alias_value":"2005.00060v2","created_at":"2026-07-05T01:15:58.642847+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.00060","created_at":"2026-07-05T01:15:58.642847+00:00"},{"alias_kind":"pith_short_12","alias_value":"MCZAZT5DFT5J","created_at":"2026-07-05T01:15:58.642847+00:00"},{"alias_kind":"pith_short_16","alias_value":"MCZAZT5DFT5JOL53","created_at":"2026-07-05T01:15:58.642847+00:00"},{"alias_kind":"pith_short_8","alias_value":"MCZAZT5D","created_at":"2026-07-05T01:15:58.642847+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2504.19239","citing_title":"The effect of the number of parameters and the number of local feature patches on loss landscapes in distributed quantum neural networks","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2506.14951","citing_title":"Flat Channels to Infinity in Neural Loss Landscapes","ref_index":51,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MCZAZT5DFT5JOL53SNUBNORC6E","json":"https://pith.science/pith/MCZAZT5DFT5JOL53SNUBNORC6E.json","graph_json":"https://pith.science/api/pith-number/MCZAZT5DFT5JOL53SNUBNORC6E/graph.json","events_json":"https://pith.science/api/pith-number/MCZAZT5DFT5JOL53SNUBNORC6E/events.json","paper":"https://pith.science/paper/MCZAZT5D"},"agent_actions":{"view_html":"https://pith.science/pith/MCZAZT5DFT5JOL53SNUBNORC6E","download_json":"https://pith.science/pith/MCZAZT5DFT5JOL53SNUBNORC6E.json","view_paper":"https://pith.science/paper/MCZAZT5D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2005.00060&json=true","fetch_graph":"https://pith.science/api/pith-number/MCZAZT5DFT5JOL53SNUBNORC6E/graph.json","fetch_events":"https://pith.science/api/pith-number/MCZAZT5DFT5JOL53SNUBNORC6E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MCZAZT5DFT5JOL53SNUBNORC6E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MCZAZT5DFT5JOL53SNUBNORC6E/action/storage_attestation","attest_author":"https://pith.science/pith/MCZAZT5DFT5JOL53SNUBNORC6E/action/author_attestation","sign_citation":"https://pith.science/pith/MCZAZT5DFT5JOL53SNUBNORC6E/action/citation_signature","submit_replication":"https://pith.science/pith/MCZAZT5DFT5JOL53SNUBNORC6E/action/replication_record"}},"created_at":"2026-07-05T01:15:58.642847+00:00","updated_at":"2026-07-05T01:15:58.642847+00:00"}