{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:WXODBQJ5EFK4ZYK7VO4JOZ42FK","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":"738440d232ce477ca8848c242733a43673cd3e5bae45b919a7196f2d1e0270d2","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-30T08:04:19Z","title_canon_sha256":"ad45fc4b76bc3db0ece2ef33e1d67fa5aa683bf4238e21e7aaad9f3e96262101"},"schema_version":"1.0","source":{"id":"2507.22457","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.22457","created_at":"2026-07-05T11:45:42Z"},{"alias_kind":"arxiv_version","alias_value":"2507.22457v1","created_at":"2026-07-05T11:45:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.22457","created_at":"2026-07-05T11:45:42Z"},{"alias_kind":"pith_short_12","alias_value":"WXODBQJ5EFK4","created_at":"2026-07-05T11:45:42Z"},{"alias_kind":"pith_short_16","alias_value":"WXODBQJ5EFK4ZYK7","created_at":"2026-07-05T11:45:42Z"},{"alias_kind":"pith_short_8","alias_value":"WXODBQJ5","created_at":"2026-07-05T11:45:42Z"}],"graph_snapshots":[{"event_id":"sha256:92b35db5f35bc81fab643715a0734f37c8d0bba432eb0e52e75750af4ab5b03f","target":"graph","created_at":"2026-07-05T11:45:42Z","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/2507.22457/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent work has argued that large language models (LLMs) are not \"abstract reasoners\", citing their poor zero-shot performance on a variety of challenging tasks as evidence. We revisit these experiments in order to add nuance to the claim. First, we show that while LLMs indeed perform poorly in a zero-shot setting, even tuning a small subset of parameters for input encoding can enable near-perfect performance. However, we also show that this finetuning does not necessarily transfer across datasets. We take this collection of empirical results as an invitation to (re-)open the discussion of wha","authors_text":"Chen Sun, Ellie Pavlick, Tian Yun","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-30T08:04:19Z","title":"What is an \"Abstract Reasoner\"? Revisiting Experiments and Arguments about Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.22457","kind":"arxiv","version":1},"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:aebdb9584b4b3c82c0829be7f969cbd8ed1c4bcba5dc5d70d2b9c099d0a56fda","target":"record","created_at":"2026-07-05T11:45:42Z","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":"738440d232ce477ca8848c242733a43673cd3e5bae45b919a7196f2d1e0270d2","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-30T08:04:19Z","title_canon_sha256":"ad45fc4b76bc3db0ece2ef33e1d67fa5aa683bf4238e21e7aaad9f3e96262101"},"schema_version":"1.0","source":{"id":"2507.22457","kind":"arxiv","version":1}},"canonical_sha256":"b5dc30c13d2155cce15fabb897679a2a892ca8d112b51d9a78e57d1e20e8a7b6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b5dc30c13d2155cce15fabb897679a2a892ca8d112b51d9a78e57d1e20e8a7b6","first_computed_at":"2026-07-05T11:45:42.792682Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:45:42.792682Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"A3yQ79Zf+opxZNBJSYVebo+fiFH3IzWdGdekI4Cbc73MZSdb5+wPIHpYH1PWNE0b5MkAvYiBxUDY19DLsWmuDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:45:42.793142Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.22457","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:aebdb9584b4b3c82c0829be7f969cbd8ed1c4bcba5dc5d70d2b9c099d0a56fda","sha256:92b35db5f35bc81fab643715a0734f37c8d0bba432eb0e52e75750af4ab5b03f"],"state_sha256":"f48adcaca28629e6b26d451c840e72871a320731d0fc3b040983f47705718af3"}