{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:TFQ7EIFLQ6XGY3AYT52XDR4COI","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":"65b2b92e738f7e6fb9dfd300074db5ca27d658f4bec83b916706a1539e9418a9","cross_cats_sorted":["cs.CL","cs.IR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-05-19T04:59:34Z","title_canon_sha256":"e9e6c7ebbf051f157e9fbe0bba21100cef5b7bca66d02596d73bb912742e3324"},"schema_version":"1.0","source":{"id":"2305.11430","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.11430","created_at":"2026-07-05T07:04:47Z"},{"alias_kind":"arxiv_version","alias_value":"2305.11430v2","created_at":"2026-07-05T07:04:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.11430","created_at":"2026-07-05T07:04:47Z"},{"alias_kind":"pith_short_12","alias_value":"TFQ7EIFLQ6XG","created_at":"2026-07-05T07:04:47Z"},{"alias_kind":"pith_short_16","alias_value":"TFQ7EIFLQ6XGY3AY","created_at":"2026-07-05T07:04:47Z"},{"alias_kind":"pith_short_8","alias_value":"TFQ7EIFL","created_at":"2026-07-05T07:04:47Z"}],"graph_snapshots":[{"event_id":"sha256:be5176f50f94cc83b82e39ae92c281f321a77128877f40ebb2d0988023b4ee3b","target":"graph","created_at":"2026-07-05T07:04:47Z","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/2305.11430/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While LLMs have shown great success in understanding and generating text in traditional conversational settings, their potential for performing ill-defined complex tasks is largely under-studied. Indeed, we are yet to conduct comprehensive benchmarking studies with multiple LLMs that are exclusively focused on a complex task. However, conducting such benchmarking studies is challenging because of the large variations in LLMs' performance when different prompt types/styles are used and different degrees of detail are provided in the prompts. To address this issue, the paper proposes a general t","authors_text":"Dongji Feng, Shubhra Kanti Karmaker Santu","cross_cats":["cs.CL","cs.IR","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-05-19T04:59:34Z","title":"TELeR: A General Taxonomy of LLM Prompts for Benchmarking Complex Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.11430","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":"sha256:8be22bc608826b50654344d0f2e2447967137003f95bf32d5f65bd9085306771","target":"record","created_at":"2026-07-05T07:04:47Z","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":"65b2b92e738f7e6fb9dfd300074db5ca27d658f4bec83b916706a1539e9418a9","cross_cats_sorted":["cs.CL","cs.IR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-05-19T04:59:34Z","title_canon_sha256":"e9e6c7ebbf051f157e9fbe0bba21100cef5b7bca66d02596d73bb912742e3324"},"schema_version":"1.0","source":{"id":"2305.11430","kind":"arxiv","version":2}},"canonical_sha256":"9961f220ab87ae6c6c189f7571c7827215d442585abaeeb7967df037f054e48d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9961f220ab87ae6c6c189f7571c7827215d442585abaeeb7967df037f054e48d","first_computed_at":"2026-07-05T07:04:47.100333Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:04:47.100333Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"H0BeOKthe38bK3dlNhFok+7c9bg1quF7zpErKM8zo8X/89G6fEkErPVCPNwqSsvYXOpnWioXxw/NsCGoNxajAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:04:47.100999Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.11430","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8be22bc608826b50654344d0f2e2447967137003f95bf32d5f65bd9085306771","sha256:be5176f50f94cc83b82e39ae92c281f321a77128877f40ebb2d0988023b4ee3b"],"state_sha256":"7c29c7d82811ebf64c537dc0606795484dfd68eccf516859dff05dd82ffd2910"}