{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PX2FHZCYNKJX6S62IHQTJSBSK5","short_pith_number":"pith:PX2FHZCY","schema_version":"1.0","canonical_sha256":"7df453e4586a937f4bda41e134c8325757c49a20d8364461033bde5b1446fc8b","source":{"kind":"arxiv","id":"2606.20933","version":1},"attestation_state":"computed","paper":{"title":"Towards Robust Training in NNGPT AutoML Pipeline: A Loss-Optimizer Pairing Selection Study","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anton Abramochkin, Dmitry Ignatov, Radu Timofte","submitted_at":"2026-06-18T20:51:42Z","abstract_excerpt":"The choice of loss function and optimizer is an important decision, that shapes further model training. Yet automated architecture search pipelines (AutoML) benefits significantly more from the optimal pairing selection and vice versa. This paper investigates whether a single recipe is sufficient for heterogeneous architecture pools, or whether the optimal pairing varies across structurally diverse models. We conduct a systematic empirical study of all $3 \\times 6 = 18$ combinations of six optimizers (SGD+Momentum, Adam, AdamW, RMSprop, Adagrad, Adadelta), paired with three loss functions: Cro"},"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":"2606.20933","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-06-18T20:51:42Z","cross_cats_sorted":[],"title_canon_sha256":"638fba496092fb211c859bb5ac2aab256fd8cfb327da8873e5bb80db3c5f2ef2","abstract_canon_sha256":"f1f4111fcbe70b29da0f433b0f7d06b783eaac0d81d21ba78195d92ebccd500a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:12:22.475541Z","signature_b64":"HEr2KXkosD0UPXyni7htnHRXWG4ckHk78khNZYWjU8dFcnG29e6qzkery8860IxR8Q1MnLVokArMX+xLlMFoAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7df453e4586a937f4bda41e134c8325757c49a20d8364461033bde5b1446fc8b","last_reissued_at":"2026-06-23T01:12:22.475120Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:12:22.475120Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Robust Training in NNGPT AutoML Pipeline: A Loss-Optimizer Pairing Selection Study","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anton Abramochkin, Dmitry Ignatov, Radu Timofte","submitted_at":"2026-06-18T20:51:42Z","abstract_excerpt":"The choice of loss function and optimizer is an important decision, that shapes further model training. Yet automated architecture search pipelines (AutoML) benefits significantly more from the optimal pairing selection and vice versa. This paper investigates whether a single recipe is sufficient for heterogeneous architecture pools, or whether the optimal pairing varies across structurally diverse models. We conduct a systematic empirical study of all $3 \\times 6 = 18$ combinations of six optimizers (SGD+Momentum, Adam, AdamW, RMSprop, Adagrad, Adadelta), paired with three loss functions: Cro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20933","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/2606.20933/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":"2606.20933","created_at":"2026-06-23T01:12:22.475190+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.20933v1","created_at":"2026-06-23T01:12:22.475190+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20933","created_at":"2026-06-23T01:12:22.475190+00:00"},{"alias_kind":"pith_short_12","alias_value":"PX2FHZCYNKJX","created_at":"2026-06-23T01:12:22.475190+00:00"},{"alias_kind":"pith_short_16","alias_value":"PX2FHZCYNKJX6S62","created_at":"2026-06-23T01:12:22.475190+00:00"},{"alias_kind":"pith_short_8","alias_value":"PX2FHZCY","created_at":"2026-06-23T01:12:22.475190+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/PX2FHZCYNKJX6S62IHQTJSBSK5","json":"https://pith.science/pith/PX2FHZCYNKJX6S62IHQTJSBSK5.json","graph_json":"https://pith.science/api/pith-number/PX2FHZCYNKJX6S62IHQTJSBSK5/graph.json","events_json":"https://pith.science/api/pith-number/PX2FHZCYNKJX6S62IHQTJSBSK5/events.json","paper":"https://pith.science/paper/PX2FHZCY"},"agent_actions":{"view_html":"https://pith.science/pith/PX2FHZCYNKJX6S62IHQTJSBSK5","download_json":"https://pith.science/pith/PX2FHZCYNKJX6S62IHQTJSBSK5.json","view_paper":"https://pith.science/paper/PX2FHZCY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.20933&json=true","fetch_graph":"https://pith.science/api/pith-number/PX2FHZCYNKJX6S62IHQTJSBSK5/graph.json","fetch_events":"https://pith.science/api/pith-number/PX2FHZCYNKJX6S62IHQTJSBSK5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PX2FHZCYNKJX6S62IHQTJSBSK5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PX2FHZCYNKJX6S62IHQTJSBSK5/action/storage_attestation","attest_author":"https://pith.science/pith/PX2FHZCYNKJX6S62IHQTJSBSK5/action/author_attestation","sign_citation":"https://pith.science/pith/PX2FHZCYNKJX6S62IHQTJSBSK5/action/citation_signature","submit_replication":"https://pith.science/pith/PX2FHZCYNKJX6S62IHQTJSBSK5/action/replication_record"}},"created_at":"2026-06-23T01:12:22.475190+00:00","updated_at":"2026-06-23T01:12:22.475190+00:00"}