{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YNXGBHPAZPK2VGCS7TO2Y4CMM7","short_pith_number":"pith:YNXGBHPA","schema_version":"1.0","canonical_sha256":"c36e609de0cbd5aa9852fcddac704c67d2f40b130d7d0eb84a22b2f094cee008","source":{"kind":"arxiv","id":"2607.00368","version":1},"attestation_state":"computed","paper":{"title":"Beyond Perplexity: A Behavioral Evaluation Framework for Deployment-Memory Claims in LLM Test-Time Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Guangyi Chen, Kun Zhang, Lingjing Kong, Shaoan Xie, Xiangchen Song, Xinshuai Dong, Zhenhao Chen","submitted_at":"2026-07-01T03:07:17Z","abstract_excerpt":"Large language model test-time training (TTT) is often evaluated through local proxy metrics: models are updated on recent tokens, retrieved context, target-domain data, or verifiable task attempts, and then judged by perplexity, future-token loss, long-context performance, or reward. These metrics are well matched to claims about stream adaptation, domain adaptation, context compression, and reward-backed test-time improvement. They are weaker evidence, however, for a capability that TTT results are increasingly used to motivate: deployed assistant memory, personalization, or sparse post-depl"},"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":"2607.00368","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-01T03:07:17Z","cross_cats_sorted":[],"title_canon_sha256":"8dc6d06ac7318c08e1052c963d238d300b36ff87a4115c812671379eb956e7cd","abstract_canon_sha256":"61f0217094fa38a883c8b9ae368c71f24fd61359504784eecddf137b62d2233b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-02T01:17:41.396632Z","signature_b64":"BDq6+yTcUe4VcY3rLqPloz8PQZk8qtRlsHr0YCb82N+IQLPuQT+5i9VDgf8Vz8VbBrcLCDaZplbNokYF5SWUCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c36e609de0cbd5aa9852fcddac704c67d2f40b130d7d0eb84a22b2f094cee008","last_reissued_at":"2026-07-02T01:17:41.396247Z","signature_status":"signed_v1","first_computed_at":"2026-07-02T01:17:41.396247Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Perplexity: A Behavioral Evaluation Framework for Deployment-Memory Claims in LLM Test-Time Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Guangyi Chen, Kun Zhang, Lingjing Kong, Shaoan Xie, Xiangchen Song, Xinshuai Dong, Zhenhao Chen","submitted_at":"2026-07-01T03:07:17Z","abstract_excerpt":"Large language model test-time training (TTT) is often evaluated through local proxy metrics: models are updated on recent tokens, retrieved context, target-domain data, or verifiable task attempts, and then judged by perplexity, future-token loss, long-context performance, or reward. These metrics are well matched to claims about stream adaptation, domain adaptation, context compression, and reward-backed test-time improvement. They are weaker evidence, however, for a capability that TTT results are increasingly used to motivate: deployed assistant memory, personalization, or sparse post-depl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.00368","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/2607.00368/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":"2607.00368","created_at":"2026-07-02T01:17:41.396307+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.00368v1","created_at":"2026-07-02T01:17:41.396307+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.00368","created_at":"2026-07-02T01:17:41.396307+00:00"},{"alias_kind":"pith_short_12","alias_value":"YNXGBHPAZPK2","created_at":"2026-07-02T01:17:41.396307+00:00"},{"alias_kind":"pith_short_16","alias_value":"YNXGBHPAZPK2VGCS","created_at":"2026-07-02T01:17:41.396307+00:00"},{"alias_kind":"pith_short_8","alias_value":"YNXGBHPA","created_at":"2026-07-02T01:17:41.396307+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/YNXGBHPAZPK2VGCS7TO2Y4CMM7","json":"https://pith.science/pith/YNXGBHPAZPK2VGCS7TO2Y4CMM7.json","graph_json":"https://pith.science/api/pith-number/YNXGBHPAZPK2VGCS7TO2Y4CMM7/graph.json","events_json":"https://pith.science/api/pith-number/YNXGBHPAZPK2VGCS7TO2Y4CMM7/events.json","paper":"https://pith.science/paper/YNXGBHPA"},"agent_actions":{"view_html":"https://pith.science/pith/YNXGBHPAZPK2VGCS7TO2Y4CMM7","download_json":"https://pith.science/pith/YNXGBHPAZPK2VGCS7TO2Y4CMM7.json","view_paper":"https://pith.science/paper/YNXGBHPA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.00368&json=true","fetch_graph":"https://pith.science/api/pith-number/YNXGBHPAZPK2VGCS7TO2Y4CMM7/graph.json","fetch_events":"https://pith.science/api/pith-number/YNXGBHPAZPK2VGCS7TO2Y4CMM7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YNXGBHPAZPK2VGCS7TO2Y4CMM7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YNXGBHPAZPK2VGCS7TO2Y4CMM7/action/storage_attestation","attest_author":"https://pith.science/pith/YNXGBHPAZPK2VGCS7TO2Y4CMM7/action/author_attestation","sign_citation":"https://pith.science/pith/YNXGBHPAZPK2VGCS7TO2Y4CMM7/action/citation_signature","submit_replication":"https://pith.science/pith/YNXGBHPAZPK2VGCS7TO2Y4CMM7/action/replication_record"}},"created_at":"2026-07-02T01:17:41.396307+00:00","updated_at":"2026-07-02T01:17:41.396307+00:00"}