{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:DZOW5YAZHPVELONQ6HGCJ6LVUV","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":"72325415a31ec0f7e64024e94fea001018a64887864c054b021cc79ba194a2a0","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T03:54:47Z","title_canon_sha256":"eba24d625903d134b280e1415f400fc13eed36f160d4db8a903eaf7730f711e1"},"schema_version":"1.0","source":{"id":"2505.19475","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19475","created_at":"2026-07-05T11:10:58Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19475v2","created_at":"2026-07-05T11:10:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19475","created_at":"2026-07-05T11:10:58Z"},{"alias_kind":"pith_short_12","alias_value":"DZOW5YAZHPVE","created_at":"2026-07-05T11:10:58Z"},{"alias_kind":"pith_short_16","alias_value":"DZOW5YAZHPVELONQ","created_at":"2026-07-05T11:10:58Z"},{"alias_kind":"pith_short_8","alias_value":"DZOW5YAZ","created_at":"2026-07-05T11:10:58Z"}],"graph_snapshots":[{"event_id":"sha256:4b59c4c5b8cd18187b65e82189a264427969c140c0edc93e49673acfb8816e7c","target":"graph","created_at":"2026-07-05T11:10:58Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2505.19475/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Learning to adapt pretrained language models to unlabeled, out-of-distribution data is a critical challenge, as models often falter on structurally novel reasoning tasks even while excelling within their training distribution. We introduce a new framework called VDS-TTT - Verifier-Driven Sample Selection for Test-Time Training to efficiently address this. We use a learned verifier to score a pool of generated responses and select only from high ranking pseudo-labeled examples for fine-tuned adaptation. Specifically, for each input query our LLM generates N candidate answers; the verifier assig","authors_text":"Hossam Amer, Mohammad Mahdi Moradi, Sudhir Mudur, Walid Ahmed, Weiwei Zhang, Yang Liu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T03:54:47Z","title":"Continuous Self-Improvement of Large Language Models by Test-time Training with Verifier-Driven Sample Selection"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19475","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a8568c0b276cb61eeed583b85be5cd8adf7a2eeb47712fa0d7cd9800d4985f18","target":"record","created_at":"2026-07-05T11:10:58Z","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":"72325415a31ec0f7e64024e94fea001018a64887864c054b021cc79ba194a2a0","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T03:54:47Z","title_canon_sha256":"eba24d625903d134b280e1415f400fc13eed36f160d4db8a903eaf7730f711e1"},"schema_version":"1.0","source":{"id":"2505.19475","kind":"arxiv","version":2}},"canonical_sha256":"1e5d6ee0193bea45b9b0f1cc24f975a546be38bbbc64b3a2e22dd54c29688b91","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1e5d6ee0193bea45b9b0f1cc24f975a546be38bbbc64b3a2e22dd54c29688b91","first_computed_at":"2026-07-05T11:10:58.213949Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:10:58.213949Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"W5NIUBnvhGkHHu/hl/VRuE2Ed/TgsrCIgUGDUdv1CL9yCdXIqgj8843PeuVGekD+syp9VuQocCLY+jFc/ERmCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:10:58.214373Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.19475","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a8568c0b276cb61eeed583b85be5cd8adf7a2eeb47712fa0d7cd9800d4985f18","sha256:4b59c4c5b8cd18187b65e82189a264427969c140c0edc93e49673acfb8816e7c"],"state_sha256":"ef2986b2407f61d365351db76884a00ef08ce25c7455f6fb8ed7cf96ed1d6b05"}