{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:SCY4MDQZT4GIN26QGWOJ4FSHSS","short_pith_number":"pith:SCY4MDQZ","canonical_record":{"source":{"id":"2511.21692","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-11-26T18:59:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cdfa46709a58ba2117f0a238743ecf2d1e6b9bbfcff73c9a17e65e1d5d6323bb","abstract_canon_sha256":"3c0d6fadc5c1cdab16642d898ba52a9621eecb9dcf6ad0699dc2ba641f188697"},"schema_version":"1.0"},"canonical_sha256":"90b1c60e199f0c86ebd0359c9e164794b2d34c9864d03ed2b191885084f753e0","source":{"kind":"arxiv","id":"2511.21692","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2511.21692","created_at":"2026-08-04T01:58:09Z"},{"alias_kind":"arxiv_version","alias_value":"2511.21692v2","created_at":"2026-08-04T01:58:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.21692","created_at":"2026-08-04T01:58:09Z"},{"alias_kind":"pith_short_12","alias_value":"SCY4MDQZT4GI","created_at":"2026-08-04T01:58:09Z"},{"alias_kind":"pith_short_16","alias_value":"SCY4MDQZT4GIN26Q","created_at":"2026-08-04T01:58:09Z"},{"alias_kind":"pith_short_8","alias_value":"SCY4MDQZ","created_at":"2026-08-04T01:58:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:SCY4MDQZT4GIN26QGWOJ4FSHSS","target":"record","payload":{"canonical_record":{"source":{"id":"2511.21692","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-11-26T18:59:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cdfa46709a58ba2117f0a238743ecf2d1e6b9bbfcff73c9a17e65e1d5d6323bb","abstract_canon_sha256":"3c0d6fadc5c1cdab16642d898ba52a9621eecb9dcf6ad0699dc2ba641f188697"},"schema_version":"1.0"},"canonical_sha256":"90b1c60e199f0c86ebd0359c9e164794b2d34c9864d03ed2b191885084f753e0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T01:58:09.681181Z","signature_b64":"KbBC48PMBC3fZdEPdVmIwnji1kK5zTQFeP8EDclRR43l2U5JvXp4GVhJK8ZbbARKe8q4rbuC5UPR6c9JgFQqAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"90b1c60e199f0c86ebd0359c9e164794b2d34c9864d03ed2b191885084f753e0","last_reissued_at":"2026-08-04T01:58:09.679531Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T01:58:09.679531Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2511.21692","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-04T01:58:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L5kuqM1FCgMGiWwFvUvC3BOjmDO2P8ch6zvovUzR2y35ciyWhaZgbzwkrWwQ56nuoMM0gwkp+TBzqqoYmn6tDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T01:36:32.039516Z"},"content_sha256":"9840e4adce7d8e42efbd3a16743be9326de5b5213c5ce492b5089107121df74b","schema_version":"1.0","event_id":"sha256:9840e4adce7d8e42efbd3a16743be9326de5b5213c5ce492b5089107121df74b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:SCY4MDQZT4GIN26QGWOJ4FSHSS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Revisiting Generalization Across Difficulty Levels: It's Not So Easy","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ilana Nguyen, Max Zuo, Nihal V. Nayak, Stephen H. Bach, Yeganeh Kordi","submitted_at":"2025-11-26T18:59:57Z","abstract_excerpt":"We investigate how well large language models (LLMs) generalize across different task difficulties, a key question for effective data curation and evaluation. Existing research is mixed regarding whether training on easier or harder data leads to better results, and whether those gains come on easier or harder test data. We address this question by conducting a systematic evaluation of LLMs' generalization across models, datasets, and fine-grained groups of example difficulty. We rank examples in six datasets using the outputs of thousands of different LLMs and Item Response Theory (IRT), a we"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.21692","kind":"arxiv","version":2},"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/2511.21692/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":"2026-08-04T01:58:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CDMdqoQE85jWSMk+2gOFqrJ59VviirsY3hDRWL7l/cyRgM652M9Xlg6XR0OXI/0qC3KsQOM33/BgikYeb8UBAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T01:36:32.040506Z"},"content_sha256":"33883d3bbfb307dadff8d74c1e10ca7b98c1537de3ea23050de399b47fb9fefe","schema_version":"1.0","event_id":"sha256:33883d3bbfb307dadff8d74c1e10ca7b98c1537de3ea23050de399b47fb9fefe"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SCY4MDQZT4GIN26QGWOJ4FSHSS/bundle.json","state_url":"https://pith.science/pith/SCY4MDQZT4GIN26QGWOJ4FSHSS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SCY4MDQZT4GIN26QGWOJ4FSHSS/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":"2026-08-09T01:36:32Z","links":{"resolver":"https://pith.science/pith/SCY4MDQZT4GIN26QGWOJ4FSHSS","bundle":"https://pith.science/pith/SCY4MDQZT4GIN26QGWOJ4FSHSS/bundle.json","state":"https://pith.science/pith/SCY4MDQZT4GIN26QGWOJ4FSHSS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SCY4MDQZT4GIN26QGWOJ4FSHSS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SCY4MDQZT4GIN26QGWOJ4FSHSS","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":"3c0d6fadc5c1cdab16642d898ba52a9621eecb9dcf6ad0699dc2ba641f188697","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-11-26T18:59:57Z","title_canon_sha256":"cdfa46709a58ba2117f0a238743ecf2d1e6b9bbfcff73c9a17e65e1d5d6323bb"},"schema_version":"1.0","source":{"id":"2511.21692","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2511.21692","created_at":"2026-08-04T01:58:09Z"},{"alias_kind":"arxiv_version","alias_value":"2511.21692v2","created_at":"2026-08-04T01:58:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.21692","created_at":"2026-08-04T01:58:09Z"},{"alias_kind":"pith_short_12","alias_value":"SCY4MDQZT4GI","created_at":"2026-08-04T01:58:09Z"},{"alias_kind":"pith_short_16","alias_value":"SCY4MDQZT4GIN26Q","created_at":"2026-08-04T01:58:09Z"},{"alias_kind":"pith_short_8","alias_value":"SCY4MDQZ","created_at":"2026-08-04T01:58:09Z"}],"graph_snapshots":[{"event_id":"sha256:33883d3bbfb307dadff8d74c1e10ca7b98c1537de3ea23050de399b47fb9fefe","target":"graph","created_at":"2026-08-04T01:58:09Z","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/2511.21692/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We investigate how well large language models (LLMs) generalize across different task difficulties, a key question for effective data curation and evaluation. Existing research is mixed regarding whether training on easier or harder data leads to better results, and whether those gains come on easier or harder test data. We address this question by conducting a systematic evaluation of LLMs' generalization across models, datasets, and fine-grained groups of example difficulty. We rank examples in six datasets using the outputs of thousands of different LLMs and Item Response Theory (IRT), a we","authors_text":"Ilana Nguyen, Max Zuo, Nihal V. Nayak, Stephen H. Bach, Yeganeh Kordi","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-11-26T18:59:57Z","title":"Revisiting Generalization Across Difficulty Levels: It's Not So Easy"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.21692","kind":"arxiv","version":2},"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":"sha256:9840e4adce7d8e42efbd3a16743be9326de5b5213c5ce492b5089107121df74b","target":"record","created_at":"2026-08-04T01:58:09Z","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":"3c0d6fadc5c1cdab16642d898ba52a9621eecb9dcf6ad0699dc2ba641f188697","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-11-26T18:59:57Z","title_canon_sha256":"cdfa46709a58ba2117f0a238743ecf2d1e6b9bbfcff73c9a17e65e1d5d6323bb"},"schema_version":"1.0","source":{"id":"2511.21692","kind":"arxiv","version":2}},"canonical_sha256":"90b1c60e199f0c86ebd0359c9e164794b2d34c9864d03ed2b191885084f753e0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"90b1c60e199f0c86ebd0359c9e164794b2d34c9864d03ed2b191885084f753e0","first_computed_at":"2026-08-04T01:58:09.679531Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-04T01:58:09.679531Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KbBC48PMBC3fZdEPdVmIwnji1kK5zTQFeP8EDclRR43l2U5JvXp4GVhJK8ZbbARKe8q4rbuC5UPR6c9JgFQqAQ==","signature_status":"signed_v1","signed_at":"2026-08-04T01:58:09.681181Z","signed_message":"canonical_sha256_bytes"},"source_id":"2511.21692","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9840e4adce7d8e42efbd3a16743be9326de5b5213c5ce492b5089107121df74b","sha256:33883d3bbfb307dadff8d74c1e10ca7b98c1537de3ea23050de399b47fb9fefe"],"state_sha256":"6c22f67ccb015cad2629ace24471036265c8dc07146003522ec884e3a0f586ce"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"16TJfczjWZN5iPHrdvsmKnW/uReruI83Mp+BWnR1XdAihHqR2on62v2179d9lgPFdaZZ+cj2DTWw+P6Zts0iAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T01:36:32.044216Z","bundle_sha256":"b2cab932e59dface3eb87fd8827e247d154fed30c53258acbf251998044600f4"}}