{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:LF2DV2RAMLCQBS73324TITBQ4D","short_pith_number":"pith:LF2DV2RA","schema_version":"1.0","canonical_sha256":"59743aea2062c500cbfbdeb9344c30e0d142ed64879e552029a98cbe60aac2d2","source":{"kind":"arxiv","id":"2508.07571","version":2},"attestation_state":"computed","paper":{"title":"Towards Theoretical Understanding of Transformer Test-Time Computing: Investigation on In-Context Linear Regression","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Beining Wu, Difan Zou, Miao Lu, Xingwu Chen","submitted_at":"2025-08-11T03:05:36Z","abstract_excerpt":"Using more test-time computation during language model inference, such as generating more intermediate thoughts or sampling multiple candidate answers, has proven effective in significantly improving model performance. This paper takes an initial step toward bridging the gap between practical language model inference and theoretical transformer analysis by incorporating randomness and sampling. We focus on in-context linear regression with continuous/binary coefficients, where our framework simulates language model decoding through noise injection and binary coefficient sampling. Through this "},"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":"2508.07571","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-11T03:05:36Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7d1645c25a55f252c2526683dd9dac36d335915e0cc10047c679dfe2658c181f","abstract_canon_sha256":"22e5d44222f5de4a0afd2444c5242dfb5a38259d14a3be9f9f1e731bcd733477"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:56:06.205392Z","signature_b64":"QmnZhiRoQUaY63KG0WnKIwjYeRHgrwHMJRzmwV6VpeJZ8yb3ugc0JJLUN/dBERrIs8UHFb6bCCNasfOxPYkACg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59743aea2062c500cbfbdeb9344c30e0d142ed64879e552029a98cbe60aac2d2","last_reissued_at":"2026-07-05T11:56:06.204896Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:56:06.204896Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Theoretical Understanding of Transformer Test-Time Computing: Investigation on In-Context Linear Regression","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Beining Wu, Difan Zou, Miao Lu, Xingwu Chen","submitted_at":"2025-08-11T03:05:36Z","abstract_excerpt":"Using more test-time computation during language model inference, such as generating more intermediate thoughts or sampling multiple candidate answers, has proven effective in significantly improving model performance. This paper takes an initial step toward bridging the gap between practical language model inference and theoretical transformer analysis by incorporating randomness and sampling. We focus on in-context linear regression with continuous/binary coefficients, where our framework simulates language model decoding through noise injection and binary coefficient sampling. Through this "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.07571","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/2508.07571/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":"2508.07571","created_at":"2026-07-05T11:56:06.204957+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.07571v2","created_at":"2026-07-05T11:56:06.204957+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07571","created_at":"2026-07-05T11:56:06.204957+00:00"},{"alias_kind":"pith_short_12","alias_value":"LF2DV2RAMLCQ","created_at":"2026-07-05T11:56:06.204957+00:00"},{"alias_kind":"pith_short_16","alias_value":"LF2DV2RAMLCQBS73","created_at":"2026-07-05T11:56:06.204957+00:00"},{"alias_kind":"pith_short_8","alias_value":"LF2DV2RA","created_at":"2026-07-05T11:56:06.204957+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.06609","citing_title":"Transformers Efficiently Perform In-Context Logistic Regression via Normalized Gradient Descent","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LF2DV2RAMLCQBS73324TITBQ4D","json":"https://pith.science/pith/LF2DV2RAMLCQBS73324TITBQ4D.json","graph_json":"https://pith.science/api/pith-number/LF2DV2RAMLCQBS73324TITBQ4D/graph.json","events_json":"https://pith.science/api/pith-number/LF2DV2RAMLCQBS73324TITBQ4D/events.json","paper":"https://pith.science/paper/LF2DV2RA"},"agent_actions":{"view_html":"https://pith.science/pith/LF2DV2RAMLCQBS73324TITBQ4D","download_json":"https://pith.science/pith/LF2DV2RAMLCQBS73324TITBQ4D.json","view_paper":"https://pith.science/paper/LF2DV2RA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.07571&json=true","fetch_graph":"https://pith.science/api/pith-number/LF2DV2RAMLCQBS73324TITBQ4D/graph.json","fetch_events":"https://pith.science/api/pith-number/LF2DV2RAMLCQBS73324TITBQ4D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LF2DV2RAMLCQBS73324TITBQ4D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LF2DV2RAMLCQBS73324TITBQ4D/action/storage_attestation","attest_author":"https://pith.science/pith/LF2DV2RAMLCQBS73324TITBQ4D/action/author_attestation","sign_citation":"https://pith.science/pith/LF2DV2RAMLCQBS73324TITBQ4D/action/citation_signature","submit_replication":"https://pith.science/pith/LF2DV2RAMLCQBS73324TITBQ4D/action/replication_record"}},"created_at":"2026-07-05T11:56:06.204957+00:00","updated_at":"2026-07-05T11:56:06.204957+00:00"}