{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FNEZ2XGCTXCJVZKOOTEINTLEHN","short_pith_number":"pith:FNEZ2XGC","schema_version":"1.0","canonical_sha256":"2b499d5cc29dc49ae54e74c886cd643b4f2e26e65aabd09b38e1646bbf79ea48","source":{"kind":"arxiv","id":"2402.12151","version":2},"attestation_state":"computed","paper":{"title":"Transformer-based Causal Language Models Perform Clustering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Lav R. Varshney, Xinbo Wu","submitted_at":"2024-02-19T14:02:31Z","abstract_excerpt":"Even though large language models (LLMs) have demonstrated remarkable capability in solving various natural language tasks, the capability of an LLM to follow human instructions is still a concern. Recent works have shown great improvements in the instruction-following capability via additional training for instruction-following tasks. However, the mechanisms responsible for effective instruction-following capabilities remain inadequately understood. Here, we introduce a simplified instruction-following task and use synthetic datasets to analyze a Transformer-based causal language model. Our f"},"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":"2402.12151","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T14:02:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cfa059b6d09cffd324a54a5edfabaac79cfca5235b3031520248c5b62188102b","abstract_canon_sha256":"fd2b1cab980a58bf79c224f04358bca633709368c518a3fb57567539d87692f2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:51:39.387331Z","signature_b64":"pTLxl2QOlZHt+Z8opvfyZedPGhmZgj4B021uBlVm8H6o77ifUKc73PysYcvPpgWMg8EAHWbZKOv4+T7pHWg8AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2b499d5cc29dc49ae54e74c886cd643b4f2e26e65aabd09b38e1646bbf79ea48","last_reissued_at":"2026-07-05T07:51:39.386868Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:51:39.386868Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Transformer-based Causal Language Models Perform Clustering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Lav R. Varshney, Xinbo Wu","submitted_at":"2024-02-19T14:02:31Z","abstract_excerpt":"Even though large language models (LLMs) have demonstrated remarkable capability in solving various natural language tasks, the capability of an LLM to follow human instructions is still a concern. Recent works have shown great improvements in the instruction-following capability via additional training for instruction-following tasks. However, the mechanisms responsible for effective instruction-following capabilities remain inadequately understood. Here, we introduce a simplified instruction-following task and use synthetic datasets to analyze a Transformer-based causal language model. Our f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12151","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/2402.12151/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":"2402.12151","created_at":"2026-07-05T07:51:39.386928+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.12151v2","created_at":"2026-07-05T07:51:39.386928+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12151","created_at":"2026-07-05T07:51:39.386928+00:00"},{"alias_kind":"pith_short_12","alias_value":"FNEZ2XGCTXCJ","created_at":"2026-07-05T07:51:39.386928+00:00"},{"alias_kind":"pith_short_16","alias_value":"FNEZ2XGCTXCJVZKO","created_at":"2026-07-05T07:51:39.386928+00:00"},{"alias_kind":"pith_short_8","alias_value":"FNEZ2XGC","created_at":"2026-07-05T07:51:39.386928+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.09627","citing_title":"Physics- and geometry-aware spatio-spectral graph neural operator for time-independent and time-dependent PDEs","ref_index":36,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FNEZ2XGCTXCJVZKOOTEINTLEHN","json":"https://pith.science/pith/FNEZ2XGCTXCJVZKOOTEINTLEHN.json","graph_json":"https://pith.science/api/pith-number/FNEZ2XGCTXCJVZKOOTEINTLEHN/graph.json","events_json":"https://pith.science/api/pith-number/FNEZ2XGCTXCJVZKOOTEINTLEHN/events.json","paper":"https://pith.science/paper/FNEZ2XGC"},"agent_actions":{"view_html":"https://pith.science/pith/FNEZ2XGCTXCJVZKOOTEINTLEHN","download_json":"https://pith.science/pith/FNEZ2XGCTXCJVZKOOTEINTLEHN.json","view_paper":"https://pith.science/paper/FNEZ2XGC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.12151&json=true","fetch_graph":"https://pith.science/api/pith-number/FNEZ2XGCTXCJVZKOOTEINTLEHN/graph.json","fetch_events":"https://pith.science/api/pith-number/FNEZ2XGCTXCJVZKOOTEINTLEHN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FNEZ2XGCTXCJVZKOOTEINTLEHN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FNEZ2XGCTXCJVZKOOTEINTLEHN/action/storage_attestation","attest_author":"https://pith.science/pith/FNEZ2XGCTXCJVZKOOTEINTLEHN/action/author_attestation","sign_citation":"https://pith.science/pith/FNEZ2XGCTXCJVZKOOTEINTLEHN/action/citation_signature","submit_replication":"https://pith.science/pith/FNEZ2XGCTXCJVZKOOTEINTLEHN/action/replication_record"}},"created_at":"2026-07-05T07:51:39.386928+00:00","updated_at":"2026-07-05T07:51:39.386928+00:00"}