{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:RAD6W7JXF3626RC6ZQCNHFC3FR","short_pith_number":"pith:RAD6W7JX","canonical_record":{"source":{"id":"2310.15147","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-23T17:52:06Z","cross_cats_sorted":[],"title_canon_sha256":"d3d1e78f027ff2fb02f0bd0e65249af19fdd0c36feec8191f08397f13cf5ce2b","abstract_canon_sha256":"26fd21071e8ff018db8bd26de8a063f3222b989d6f590aed211bdf4e4213ffab"},"schema_version":"1.0"},"canonical_sha256":"8807eb7d372efdaf445ecc04d3945b2c5992fb12e395581b847969ef9b189544","source":{"kind":"arxiv","id":"2310.15147","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.15147","created_at":"2026-07-05T08:05:04Z"},{"alias_kind":"arxiv_version","alias_value":"2310.15147v2","created_at":"2026-07-05T08:05:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.15147","created_at":"2026-07-05T08:05:04Z"},{"alias_kind":"pith_short_12","alias_value":"RAD6W7JXF362","created_at":"2026-07-05T08:05:04Z"},{"alias_kind":"pith_short_16","alias_value":"RAD6W7JXF3626RC6","created_at":"2026-07-05T08:05:04Z"},{"alias_kind":"pith_short_8","alias_value":"RAD6W7JX","created_at":"2026-07-05T08:05:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:RAD6W7JXF3626RC6ZQCNHFC3FR","target":"record","payload":{"canonical_record":{"source":{"id":"2310.15147","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-23T17:52:06Z","cross_cats_sorted":[],"title_canon_sha256":"d3d1e78f027ff2fb02f0bd0e65249af19fdd0c36feec8191f08397f13cf5ce2b","abstract_canon_sha256":"26fd21071e8ff018db8bd26de8a063f3222b989d6f590aed211bdf4e4213ffab"},"schema_version":"1.0"},"canonical_sha256":"8807eb7d372efdaf445ecc04d3945b2c5992fb12e395581b847969ef9b189544","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:05:04.249888Z","signature_b64":"3uvacn32jdjfgfSG0H55r8up/no+gwfd+3f72ViuQyJ4BXe/WH/QFPBj/P/UnTvy0ZJ7iGa368fU8YB4Pb45CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8807eb7d372efdaf445ecc04d3945b2c5992fb12e395581b847969ef9b189544","last_reissued_at":"2026-07-05T08:05:04.249400Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:05:04.249400Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2310.15147","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":"2026-07-05T08:05:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YtrmEBVNtieI05JP67xIfNUi3dD79dJQSLDhwCFdW+iq6n2FjO/KbBGaNKZxCnmF+ZhTl9qlnr7Odfe/5eXfDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T22:45:49.610554Z"},"content_sha256":"5cb86fbd4c894854e946e5fdc24f7642315f5f6d7ec286b29fb428953f739230","schema_version":"1.0","event_id":"sha256:5cb86fbd4c894854e946e5fdc24f7642315f5f6d7ec286b29fb428953f739230"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:RAD6W7JXF3626RC6ZQCNHFC3FR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"S3Eval: A Synthetic, Scalable, Systematic Evaluation Suite for Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fangyu Lei, Jun Zhao, Kang Liu, Qian Liu, Shizhu He, Yiming Huang","submitted_at":"2023-10-23T17:52:06Z","abstract_excerpt":"The rapid development of Large Language Models (LLMs) has led to great strides in model capabilities like long-context understanding and reasoning. However, as LLMs are able to process longer contexts, it becomes more challenging to evaluate whether they have acquired certain capabilities, since the length of text (e.g., 200K tokens) they can process far exceeds what humans can reliably assess in a reasonable duration. In this paper, we propose using complex synthetic tasks as a proxy evaluation method, and present S3Eval, a Synthetic, Scalable, Systematic evaluation suite for LLMs evaluation."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.15147","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/2310.15147/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":"2026-07-05T08:05:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1GJz7NQyHi27oKMb94VRcamL/Di+H1dYkJY3zJqMv+jjkJ/MDSuVuLDv8fE6U4CZL8mZGb/hE5ArNvjtCVWmDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T22:45:49.611306Z"},"content_sha256":"f91b0d856590ce067a102ee33d423908418a9d10c5e7f17ad0e2de4f89eb6680","schema_version":"1.0","event_id":"sha256:f91b0d856590ce067a102ee33d423908418a9d10c5e7f17ad0e2de4f89eb6680"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RAD6W7JXF3626RC6ZQCNHFC3FR/bundle.json","state_url":"https://pith.science/pith/RAD6W7JXF3626RC6ZQCNHFC3FR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RAD6W7JXF3626RC6ZQCNHFC3FR/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":"2026-08-08T22:45:49Z","links":{"resolver":"https://pith.science/pith/RAD6W7JXF3626RC6ZQCNHFC3FR","bundle":"https://pith.science/pith/RAD6W7JXF3626RC6ZQCNHFC3FR/bundle.json","state":"https://pith.science/pith/RAD6W7JXF3626RC6ZQCNHFC3FR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RAD6W7JXF3626RC6ZQCNHFC3FR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:RAD6W7JXF3626RC6ZQCNHFC3FR","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":"26fd21071e8ff018db8bd26de8a063f3222b989d6f590aed211bdf4e4213ffab","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-23T17:52:06Z","title_canon_sha256":"d3d1e78f027ff2fb02f0bd0e65249af19fdd0c36feec8191f08397f13cf5ce2b"},"schema_version":"1.0","source":{"id":"2310.15147","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.15147","created_at":"2026-07-05T08:05:04Z"},{"alias_kind":"arxiv_version","alias_value":"2310.15147v2","created_at":"2026-07-05T08:05:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.15147","created_at":"2026-07-05T08:05:04Z"},{"alias_kind":"pith_short_12","alias_value":"RAD6W7JXF362","created_at":"2026-07-05T08:05:04Z"},{"alias_kind":"pith_short_16","alias_value":"RAD6W7JXF3626RC6","created_at":"2026-07-05T08:05:04Z"},{"alias_kind":"pith_short_8","alias_value":"RAD6W7JX","created_at":"2026-07-05T08:05:04Z"}],"graph_snapshots":[{"event_id":"sha256:f91b0d856590ce067a102ee33d423908418a9d10c5e7f17ad0e2de4f89eb6680","target":"graph","created_at":"2026-07-05T08:05:04Z","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/2310.15147/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The rapid development of Large Language Models (LLMs) has led to great strides in model capabilities like long-context understanding and reasoning. However, as LLMs are able to process longer contexts, it becomes more challenging to evaluate whether they have acquired certain capabilities, since the length of text (e.g., 200K tokens) they can process far exceeds what humans can reliably assess in a reasonable duration. In this paper, we propose using complex synthetic tasks as a proxy evaluation method, and present S3Eval, a Synthetic, Scalable, Systematic evaluation suite for LLMs evaluation.","authors_text":"Fangyu Lei, Jun Zhao, Kang Liu, Qian Liu, Shizhu He, Yiming Huang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-23T17:52:06Z","title":"S3Eval: A Synthetic, Scalable, Systematic Evaluation Suite for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.15147","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":"sha256:5cb86fbd4c894854e946e5fdc24f7642315f5f6d7ec286b29fb428953f739230","target":"record","created_at":"2026-07-05T08:05:04Z","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":"26fd21071e8ff018db8bd26de8a063f3222b989d6f590aed211bdf4e4213ffab","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-23T17:52:06Z","title_canon_sha256":"d3d1e78f027ff2fb02f0bd0e65249af19fdd0c36feec8191f08397f13cf5ce2b"},"schema_version":"1.0","source":{"id":"2310.15147","kind":"arxiv","version":2}},"canonical_sha256":"8807eb7d372efdaf445ecc04d3945b2c5992fb12e395581b847969ef9b189544","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8807eb7d372efdaf445ecc04d3945b2c5992fb12e395581b847969ef9b189544","first_computed_at":"2026-07-05T08:05:04.249400Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:05:04.249400Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3uvacn32jdjfgfSG0H55r8up/no+gwfd+3f72ViuQyJ4BXe/WH/QFPBj/P/UnTvy0ZJ7iGa368fU8YB4Pb45CQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:05:04.249888Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.15147","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5cb86fbd4c894854e946e5fdc24f7642315f5f6d7ec286b29fb428953f739230","sha256:f91b0d856590ce067a102ee33d423908418a9d10c5e7f17ad0e2de4f89eb6680"],"state_sha256":"f0e33e72df959bf1be3dfabc9a36d17755e6cf2ccc0b65f73bf2febc62d89245"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1lwm4cCOrjDcc+km+gcJYtq1ebucyxbveUZ4/WSHHbf0h8TDEkk8CyCtuVaBsDi5K8lG1DjxmEUe6uGe7GBHDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T22:45:49.617620Z","bundle_sha256":"7a929bd1824bd4f989dc75ea4febdda3ef96161007a1fbc95036f4bc15ca99c0"}}