{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:6KHS4CL4MSCZGIY3GYRXTLZEHC","short_pith_number":"pith:6KHS4CL4","schema_version":"1.0","canonical_sha256":"f28f2e097c648593231b362379af2438a79a71026c4aa5a160e501611e6fded2","source":{"kind":"arxiv","id":"2406.04229","version":1},"attestation_state":"computed","paper":{"title":"The CLRS-Text Algorithmic Reasoning Language Benchmark","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.DS","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alex Vitvitskyi, Avi Schwarzschild, Borja Ibarz, Charles Blundell, Larisa Markeeva, Olga Kozlova, Petar Veli\\v{c}kovi\\'c, Sean McLeish, Tom Goldstein, Wilfried Bounsi","submitted_at":"2024-06-06T16:29:25Z","abstract_excerpt":"Eliciting reasoning capabilities from language models (LMs) is a critical direction on the path towards building intelligent systems. Most recent studies dedicated to reasoning focus on out-of-distribution performance on procedurally-generated synthetic benchmarks, bespoke-built to evaluate specific skills only. This trend makes results hard to transfer across publications, slowing down progress. Three years ago, a similar issue was identified and rectified in the field of neural algorithmic reasoning, with the advent of the CLRS benchmark. CLRS is a dataset generator comprising graph executio"},"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":"2406.04229","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-06T16:29:25Z","cross_cats_sorted":["cs.AI","cs.CL","cs.DS","stat.ML"],"title_canon_sha256":"6383462608e7006aa82b5ff70b3238421faa9ea2bc169fb4f371a81f138e5d53","abstract_canon_sha256":"056fcb335094a2f257eaa1fb10b698d1ba8784df725d44a1b826fde36c212496"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:28.257713Z","signature_b64":"VtHIvu+0RiQpU8ots72eOW5DmsF6jsgXJ2OjBvW73IR0efH2SA4bJER2lgjHcFdVrYcUxPghP/1rVHTHAqn6CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f28f2e097c648593231b362379af2438a79a71026c4aa5a160e501611e6fded2","last_reissued_at":"2026-07-05T08:28:28.257231Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:28.257231Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The CLRS-Text Algorithmic Reasoning Language Benchmark","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.DS","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alex Vitvitskyi, Avi Schwarzschild, Borja Ibarz, Charles Blundell, Larisa Markeeva, Olga Kozlova, Petar Veli\\v{c}kovi\\'c, Sean McLeish, Tom Goldstein, Wilfried Bounsi","submitted_at":"2024-06-06T16:29:25Z","abstract_excerpt":"Eliciting reasoning capabilities from language models (LMs) is a critical direction on the path towards building intelligent systems. Most recent studies dedicated to reasoning focus on out-of-distribution performance on procedurally-generated synthetic benchmarks, bespoke-built to evaluate specific skills only. This trend makes results hard to transfer across publications, slowing down progress. Three years ago, a similar issue was identified and rectified in the field of neural algorithmic reasoning, with the advent of the CLRS benchmark. CLRS is a dataset generator comprising graph executio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.04229","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/2406.04229/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":"2406.04229","created_at":"2026-07-05T08:28:28.257290+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.04229v1","created_at":"2026-07-05T08:28:28.257290+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.04229","created_at":"2026-07-05T08:28:28.257290+00:00"},{"alias_kind":"pith_short_12","alias_value":"6KHS4CL4MSCZ","created_at":"2026-07-05T08:28:28.257290+00:00"},{"alias_kind":"pith_short_16","alias_value":"6KHS4CL4MSCZGIY3","created_at":"2026-07-05T08:28:28.257290+00:00"},{"alias_kind":"pith_short_8","alias_value":"6KHS4CL4","created_at":"2026-07-05T08:28:28.257290+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24980","citing_title":"Closed-Loop Graph Algorithm Execution with Small Language Models: Step Accuracy and Rollout Reliability","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2410.05229","citing_title":"GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2502.05171","citing_title":"Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach","ref_index":107,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6KHS4CL4MSCZGIY3GYRXTLZEHC","json":"https://pith.science/pith/6KHS4CL4MSCZGIY3GYRXTLZEHC.json","graph_json":"https://pith.science/api/pith-number/6KHS4CL4MSCZGIY3GYRXTLZEHC/graph.json","events_json":"https://pith.science/api/pith-number/6KHS4CL4MSCZGIY3GYRXTLZEHC/events.json","paper":"https://pith.science/paper/6KHS4CL4"},"agent_actions":{"view_html":"https://pith.science/pith/6KHS4CL4MSCZGIY3GYRXTLZEHC","download_json":"https://pith.science/pith/6KHS4CL4MSCZGIY3GYRXTLZEHC.json","view_paper":"https://pith.science/paper/6KHS4CL4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.04229&json=true","fetch_graph":"https://pith.science/api/pith-number/6KHS4CL4MSCZGIY3GYRXTLZEHC/graph.json","fetch_events":"https://pith.science/api/pith-number/6KHS4CL4MSCZGIY3GYRXTLZEHC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6KHS4CL4MSCZGIY3GYRXTLZEHC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6KHS4CL4MSCZGIY3GYRXTLZEHC/action/storage_attestation","attest_author":"https://pith.science/pith/6KHS4CL4MSCZGIY3GYRXTLZEHC/action/author_attestation","sign_citation":"https://pith.science/pith/6KHS4CL4MSCZGIY3GYRXTLZEHC/action/citation_signature","submit_replication":"https://pith.science/pith/6KHS4CL4MSCZGIY3GYRXTLZEHC/action/replication_record"}},"created_at":"2026-07-05T08:28:28.257290+00:00","updated_at":"2026-07-05T08:28:28.257290+00:00"}