{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HFQWCDAXMEQTFLJS74LJCC6UZC","short_pith_number":"pith:HFQWCDAX","schema_version":"1.0","canonical_sha256":"3961610c17612132ad32ff16910bd4c8bdfcd6275a13e5dd3bfd1043a7c6baa8","source":{"kind":"arxiv","id":"2410.07866","version":5},"attestation_state":"computed","paper":{"title":"System 2 Reasoning for Human-AI Alignment: Generality and Adaptivity via ARC-AGI","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Sejin Kim, Sundong Kim","submitted_at":"2024-10-10T12:34:25Z","abstract_excerpt":"Despite their broad applicability, transformer-based models still fall short in System~2 reasoning, lacking the generality and adaptivity needed for human--AI alignment. We examine weaknesses on ARC-AGI tasks, revealing gaps in compositional generalization and novel-rule adaptation, and argue that closing these gaps requires overhauling the reasoning pipeline and its evaluation. We propose three research axes: (1) Symbolic representation pipeline for compositional generality, (2) Interactive feedback-driven reasoning loop for adaptivity, and (3) Test-time task augmentation balancing both quali"},"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":"2410.07866","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-10-10T12:34:25Z","cross_cats_sorted":[],"title_canon_sha256":"9c6f77b06f20c9f34da348db50c80c8c7d5da4711954794f8c8b3c9227f9671e","abstract_canon_sha256":"e18f6e92e8dc2d93fe194233aac68d7864827b48f8afe8042e5ab1bc049deb91"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:52:57.077826Z","signature_b64":"iIA2K4rvQPvYtxHsS1j7cXwzzHR07LOfDNzqRH8XUjZZRCX5pG44kyN++kNuzANjrpn57tAGK5psdXDsQ+h/CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3961610c17612132ad32ff16910bd4c8bdfcd6275a13e5dd3bfd1043a7c6baa8","last_reissued_at":"2026-07-05T11:52:57.077284Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:52:57.077284Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"System 2 Reasoning for Human-AI Alignment: Generality and Adaptivity via ARC-AGI","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Sejin Kim, Sundong Kim","submitted_at":"2024-10-10T12:34:25Z","abstract_excerpt":"Despite their broad applicability, transformer-based models still fall short in System~2 reasoning, lacking the generality and adaptivity needed for human--AI alignment. We examine weaknesses on ARC-AGI tasks, revealing gaps in compositional generalization and novel-rule adaptation, and argue that closing these gaps requires overhauling the reasoning pipeline and its evaluation. We propose three research axes: (1) Symbolic representation pipeline for compositional generality, (2) Interactive feedback-driven reasoning loop for adaptivity, and (3) Test-time task augmentation balancing both quali"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.07866","kind":"arxiv","version":5},"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/2410.07866/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":"2410.07866","created_at":"2026-07-05T11:52:57.077353+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.07866v5","created_at":"2026-07-05T11:52:57.077353+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.07866","created_at":"2026-07-05T11:52:57.077353+00:00"},{"alias_kind":"pith_short_12","alias_value":"HFQWCDAXMEQT","created_at":"2026-07-05T11:52:57.077353+00:00"},{"alias_kind":"pith_short_16","alias_value":"HFQWCDAXMEQTFLJS","created_at":"2026-07-05T11:52:57.077353+00:00"},{"alias_kind":"pith_short_8","alias_value":"HFQWCDAX","created_at":"2026-07-05T11:52:57.077353+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27981","citing_title":"ToxiREX: A Dataset on Toxic REasoning in ConteXt","ref_index":218,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HFQWCDAXMEQTFLJS74LJCC6UZC","json":"https://pith.science/pith/HFQWCDAXMEQTFLJS74LJCC6UZC.json","graph_json":"https://pith.science/api/pith-number/HFQWCDAXMEQTFLJS74LJCC6UZC/graph.json","events_json":"https://pith.science/api/pith-number/HFQWCDAXMEQTFLJS74LJCC6UZC/events.json","paper":"https://pith.science/paper/HFQWCDAX"},"agent_actions":{"view_html":"https://pith.science/pith/HFQWCDAXMEQTFLJS74LJCC6UZC","download_json":"https://pith.science/pith/HFQWCDAXMEQTFLJS74LJCC6UZC.json","view_paper":"https://pith.science/paper/HFQWCDAX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.07866&json=true","fetch_graph":"https://pith.science/api/pith-number/HFQWCDAXMEQTFLJS74LJCC6UZC/graph.json","fetch_events":"https://pith.science/api/pith-number/HFQWCDAXMEQTFLJS74LJCC6UZC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HFQWCDAXMEQTFLJS74LJCC6UZC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HFQWCDAXMEQTFLJS74LJCC6UZC/action/storage_attestation","attest_author":"https://pith.science/pith/HFQWCDAXMEQTFLJS74LJCC6UZC/action/author_attestation","sign_citation":"https://pith.science/pith/HFQWCDAXMEQTFLJS74LJCC6UZC/action/citation_signature","submit_replication":"https://pith.science/pith/HFQWCDAXMEQTFLJS74LJCC6UZC/action/replication_record"}},"created_at":"2026-07-05T11:52:57.077353+00:00","updated_at":"2026-07-05T11:52:57.077353+00:00"}