{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QC4FNFRY5ZWUFTOF5NJW4G6VAK","short_pith_number":"pith:QC4FNFRY","schema_version":"1.0","canonical_sha256":"80b8569638ee6d42cdc5eb536e1bd5028f0498fc576ca4e1a214b567dbf5c45a","source":{"kind":"arxiv","id":"2502.08946","version":1},"attestation_state":"computed","paper":{"title":"The Stochastic Parrot on LLM's Shoulder: A Summative Assessment of Physical Concept Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.CL","authors_text":"Dit-Yan Yeung, Jiangnan Li, Jie Zhou, Junjie Wu, Lemao Liu, Mo Yu, Shunchi Zhang, Tsz Ting Chung","submitted_at":"2025-02-13T04:00:03Z","abstract_excerpt":"In a systematic way, we investigate a widely asked question: Do LLMs really understand what they say?, which relates to the more familiar term Stochastic Parrot. To this end, we propose a summative assessment over a carefully designed physical concept understanding task, PhysiCo. Our task alleviates the memorization issue via the usage of grid-format inputs that abstractly describe physical phenomena. The grids represents varying levels of understanding, from the core phenomenon, application examples to analogies to other abstract patterns in the grid world. A comprehensive study on our task d"},"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":"2502.08946","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-13T04:00:03Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"title_canon_sha256":"0c3a99450c6c578c8911fbe9fc7d6c6434ad21f74719abfa1068eca30ae87033","abstract_canon_sha256":"3fbe7fa8f46cb0f5cda9aa083308bba4844c465b922e75a95db3dd388e8eacbc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:50.216754Z","signature_b64":"TVLv2IMb3WC8xFk/CximG1nxP5n1UpHQbN/gmtXenzPcza88fUIDhf+zWNmMw4/EtDg/9EIFjcSMQkH/bBPPCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"80b8569638ee6d42cdc5eb536e1bd5028f0498fc576ca4e1a214b567dbf5c45a","last_reissued_at":"2026-07-05T10:13:50.216260Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:50.216260Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Stochastic Parrot on LLM's Shoulder: A Summative Assessment of Physical Concept Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.CL","authors_text":"Dit-Yan Yeung, Jiangnan Li, Jie Zhou, Junjie Wu, Lemao Liu, Mo Yu, Shunchi Zhang, Tsz Ting Chung","submitted_at":"2025-02-13T04:00:03Z","abstract_excerpt":"In a systematic way, we investigate a widely asked question: Do LLMs really understand what they say?, which relates to the more familiar term Stochastic Parrot. To this end, we propose a summative assessment over a carefully designed physical concept understanding task, PhysiCo. Our task alleviates the memorization issue via the usage of grid-format inputs that abstractly describe physical phenomena. The grids represents varying levels of understanding, from the core phenomenon, application examples to analogies to other abstract patterns in the grid world. A comprehensive study on our task d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.08946","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/2502.08946/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":"2502.08946","created_at":"2026-07-05T10:13:50.216313+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.08946v1","created_at":"2026-07-05T10:13:50.216313+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.08946","created_at":"2026-07-05T10:13:50.216313+00:00"},{"alias_kind":"pith_short_12","alias_value":"QC4FNFRY5ZWU","created_at":"2026-07-05T10:13:50.216313+00:00"},{"alias_kind":"pith_short_16","alias_value":"QC4FNFRY5ZWUFTOF","created_at":"2026-07-05T10:13:50.216313+00:00"},{"alias_kind":"pith_short_8","alias_value":"QC4FNFRY","created_at":"2026-07-05T10:13:50.216313+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/QC4FNFRY5ZWUFTOF5NJW4G6VAK","json":"https://pith.science/pith/QC4FNFRY5ZWUFTOF5NJW4G6VAK.json","graph_json":"https://pith.science/api/pith-number/QC4FNFRY5ZWUFTOF5NJW4G6VAK/graph.json","events_json":"https://pith.science/api/pith-number/QC4FNFRY5ZWUFTOF5NJW4G6VAK/events.json","paper":"https://pith.science/paper/QC4FNFRY"},"agent_actions":{"view_html":"https://pith.science/pith/QC4FNFRY5ZWUFTOF5NJW4G6VAK","download_json":"https://pith.science/pith/QC4FNFRY5ZWUFTOF5NJW4G6VAK.json","view_paper":"https://pith.science/paper/QC4FNFRY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.08946&json=true","fetch_graph":"https://pith.science/api/pith-number/QC4FNFRY5ZWUFTOF5NJW4G6VAK/graph.json","fetch_events":"https://pith.science/api/pith-number/QC4FNFRY5ZWUFTOF5NJW4G6VAK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QC4FNFRY5ZWUFTOF5NJW4G6VAK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QC4FNFRY5ZWUFTOF5NJW4G6VAK/action/storage_attestation","attest_author":"https://pith.science/pith/QC4FNFRY5ZWUFTOF5NJW4G6VAK/action/author_attestation","sign_citation":"https://pith.science/pith/QC4FNFRY5ZWUFTOF5NJW4G6VAK/action/citation_signature","submit_replication":"https://pith.science/pith/QC4FNFRY5ZWUFTOF5NJW4G6VAK/action/replication_record"}},"created_at":"2026-07-05T10:13:50.216313+00:00","updated_at":"2026-07-05T10:13:50.216313+00:00"}