{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:RBRQS3LK3S5YNHGMFLUOXD2D5M","short_pith_number":"pith:RBRQS3LK","canonical_record":{"source":{"id":"2505.01563","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-02T20:03:21Z","cross_cats_sorted":[],"title_canon_sha256":"f308f03a05ec0dfed3136e104ee2c08855d436ce92f03b9d02bd6402aec71638","abstract_canon_sha256":"7e2b33e301d44a18c55b01c1ed25f671b2345cc32bcbfa77d81c401025173ca7"},"schema_version":"1.0"},"canonical_sha256":"8863096d6adcbb869ccc2ae8eb8f43eb04b6a19378eaf81de156c5c20956be08","source":{"kind":"arxiv","id":"2505.01563","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.01563","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"arxiv_version","alias_value":"2505.01563v1","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.01563","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"pith_short_12","alias_value":"RBRQS3LK3S5Y","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"pith_short_16","alias_value":"RBRQS3LK3S5YNHGM","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"pith_short_8","alias_value":"RBRQS3LK","created_at":"2026-07-05T10:58:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:RBRQS3LK3S5YNHGMFLUOXD2D5M","target":"record","payload":{"canonical_record":{"source":{"id":"2505.01563","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-02T20:03:21Z","cross_cats_sorted":[],"title_canon_sha256":"f308f03a05ec0dfed3136e104ee2c08855d436ce92f03b9d02bd6402aec71638","abstract_canon_sha256":"7e2b33e301d44a18c55b01c1ed25f671b2345cc32bcbfa77d81c401025173ca7"},"schema_version":"1.0"},"canonical_sha256":"8863096d6adcbb869ccc2ae8eb8f43eb04b6a19378eaf81de156c5c20956be08","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:58:00.033870Z","signature_b64":"1gMOMkY3ON6nDTp2/Jzv92dvpz6jMl2HFOwFJZ2eVLnUOOl1C9bWTFZT14Vt9cZdLnuz/5tAmyRspg29PgxSAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8863096d6adcbb869ccc2ae8eb8f43eb04b6a19378eaf81de156c5c20956be08","last_reissued_at":"2026-07-05T10:58:00.033390Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:58:00.033390Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.01563","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:58:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DaftSn+NjBQbQzGsZHi0XymdSLKChsZsQW7lE/SwLWYvmdfs+bvjlffzRAyKpLcBxHw5OsXMp6Sy2S08K9dwDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T05:12:10.835473Z"},"content_sha256":"81d6fe09cedc5f9c408ee8ed6477bb8b5edc34770291192c755d58c98d8f18fc","schema_version":"1.0","event_id":"sha256:81d6fe09cedc5f9c408ee8ed6477bb8b5edc34770291192c755d58c98d8f18fc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:RBRQS3LK3S5YNHGMFLUOXD2D5M","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TutorGym: A Testbed for Evaluating AI Agents as Tutors and Students","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Christopher J. MacLellan, Daniel Weitekamp, Momin N. Siddiqui","submitted_at":"2025-05-02T20:03:21Z","abstract_excerpt":"Recent improvements in large language model (LLM) performance on academic benchmarks, such as MATH and GSM8K, have emboldened their use as standalone tutors and as simulations of human learning. However, these new applications require more than evaluations of final solution generation. We introduce TutorGym to evaluate these applications more directly. TutorGym is a standard interface for testing artificial intelligence (AI) agents within existing intelligent tutoring systems (ITS) that have been tested and refined in classroom studies, including Cognitive Tutors (CTAT), Apprentice Tutors, and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.01563","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/2505.01563/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:58:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"z0qQF295yUHto/85AUS60PxsdclW9Is88F+TGp5yO+A1a7eaSQmlDVUfBTVMAFpxm7ADQ2l4apxPDajDmTHKCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T05:12:10.835993Z"},"content_sha256":"09379b1e826a602b4693ee58a75ac9d2bfe2c1445446f6abe6fac31db35c8e54","schema_version":"1.0","event_id":"sha256:09379b1e826a602b4693ee58a75ac9d2bfe2c1445446f6abe6fac31db35c8e54"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RBRQS3LK3S5YNHGMFLUOXD2D5M/bundle.json","state_url":"https://pith.science/pith/RBRQS3LK3S5YNHGMFLUOXD2D5M/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RBRQS3LK3S5YNHGMFLUOXD2D5M/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-19T05:12:10Z","links":{"resolver":"https://pith.science/pith/RBRQS3LK3S5YNHGMFLUOXD2D5M","bundle":"https://pith.science/pith/RBRQS3LK3S5YNHGMFLUOXD2D5M/bundle.json","state":"https://pith.science/pith/RBRQS3LK3S5YNHGMFLUOXD2D5M/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RBRQS3LK3S5YNHGMFLUOXD2D5M/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:RBRQS3LK3S5YNHGMFLUOXD2D5M","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":"7e2b33e301d44a18c55b01c1ed25f671b2345cc32bcbfa77d81c401025173ca7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-02T20:03:21Z","title_canon_sha256":"f308f03a05ec0dfed3136e104ee2c08855d436ce92f03b9d02bd6402aec71638"},"schema_version":"1.0","source":{"id":"2505.01563","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.01563","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"arxiv_version","alias_value":"2505.01563v1","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.01563","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"pith_short_12","alias_value":"RBRQS3LK3S5Y","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"pith_short_16","alias_value":"RBRQS3LK3S5YNHGM","created_at":"2026-07-05T10:58:00Z"},{"alias_kind":"pith_short_8","alias_value":"RBRQS3LK","created_at":"2026-07-05T10:58:00Z"}],"graph_snapshots":[{"event_id":"sha256:09379b1e826a602b4693ee58a75ac9d2bfe2c1445446f6abe6fac31db35c8e54","target":"graph","created_at":"2026-07-05T10:58:00Z","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/2505.01563/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent improvements in large language model (LLM) performance on academic benchmarks, such as MATH and GSM8K, have emboldened their use as standalone tutors and as simulations of human learning. However, these new applications require more than evaluations of final solution generation. We introduce TutorGym to evaluate these applications more directly. TutorGym is a standard interface for testing artificial intelligence (AI) agents within existing intelligent tutoring systems (ITS) that have been tested and refined in classroom studies, including Cognitive Tutors (CTAT), Apprentice Tutors, and","authors_text":"Christopher J. MacLellan, Daniel Weitekamp, Momin N. Siddiqui","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-02T20:03:21Z","title":"TutorGym: A Testbed for Evaluating AI Agents as Tutors and Students"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.01563","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":"sha256:81d6fe09cedc5f9c408ee8ed6477bb8b5edc34770291192c755d58c98d8f18fc","target":"record","created_at":"2026-07-05T10:58:00Z","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":"7e2b33e301d44a18c55b01c1ed25f671b2345cc32bcbfa77d81c401025173ca7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-02T20:03:21Z","title_canon_sha256":"f308f03a05ec0dfed3136e104ee2c08855d436ce92f03b9d02bd6402aec71638"},"schema_version":"1.0","source":{"id":"2505.01563","kind":"arxiv","version":1}},"canonical_sha256":"8863096d6adcbb869ccc2ae8eb8f43eb04b6a19378eaf81de156c5c20956be08","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8863096d6adcbb869ccc2ae8eb8f43eb04b6a19378eaf81de156c5c20956be08","first_computed_at":"2026-07-05T10:58:00.033390Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:58:00.033390Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1gMOMkY3ON6nDTp2/Jzv92dvpz6jMl2HFOwFJZ2eVLnUOOl1C9bWTFZT14Vt9cZdLnuz/5tAmyRspg29PgxSAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:58:00.033870Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.01563","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:81d6fe09cedc5f9c408ee8ed6477bb8b5edc34770291192c755d58c98d8f18fc","sha256:09379b1e826a602b4693ee58a75ac9d2bfe2c1445446f6abe6fac31db35c8e54"],"state_sha256":"5373989b37f9009066cca8bedcf30c8b8c1c8e9d0f2d47dd953cb0cb94faac50"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4oQ/aX9ntgPugaafZCc9E0qPehNC61S7A0UyyAqNhbYfBFD5zl0nWy3pVBsXmdCTa74jN2qmiomAfOoyBbf/Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T05:12:10.839951Z","bundle_sha256":"dd41e5ee58db2d0fa52e29e46a53d37c98f01edfce6737db9b96633bdf5bac7e"}}