{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:Y4MJW4VZPPQ3WV26HSZHYRI7XO","short_pith_number":"pith:Y4MJW4VZ","canonical_record":{"source":{"id":"2212.10711","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-20T18:35:33Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6a5f3abe1f0c1079a45f9c8eea70c5ccd6e7625987854dc924b66f72eecc5ee1","abstract_canon_sha256":"dbe322587788b5922a38d1c1a9253493885ec221b786f350bc070378eb5065ad"},"schema_version":"1.0"},"canonical_sha256":"c7189b72b97be1bb575e3cb27c451fbba2355580e99a0cdecca08e535b8917e4","source":{"kind":"arxiv","id":"2212.10711","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2212.10711","created_at":"2026-07-05T05:27:22Z"},{"alias_kind":"arxiv_version","alias_value":"2212.10711v1","created_at":"2026-07-05T05:27:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.10711","created_at":"2026-07-05T05:27:22Z"},{"alias_kind":"pith_short_12","alias_value":"Y4MJW4VZPPQ3","created_at":"2026-07-05T05:27:22Z"},{"alias_kind":"pith_short_16","alias_value":"Y4MJW4VZPPQ3WV26","created_at":"2026-07-05T05:27:22Z"},{"alias_kind":"pith_short_8","alias_value":"Y4MJW4VZ","created_at":"2026-07-05T05:27:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:Y4MJW4VZPPQ3WV26HSZHYRI7XO","target":"record","payload":{"canonical_record":{"source":{"id":"2212.10711","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-20T18:35:33Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6a5f3abe1f0c1079a45f9c8eea70c5ccd6e7625987854dc924b66f72eecc5ee1","abstract_canon_sha256":"dbe322587788b5922a38d1c1a9253493885ec221b786f350bc070378eb5065ad"},"schema_version":"1.0"},"canonical_sha256":"c7189b72b97be1bb575e3cb27c451fbba2355580e99a0cdecca08e535b8917e4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:27:22.099075Z","signature_b64":"qmVxQOxt6Mpz/u9WMjPgPM+nVfEmuosOjRi0OkCcdMc+Fb3lMevGgWkvt+N4jijqR3ZL7cMlxnpwTf0EXoP/CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c7189b72b97be1bb575e3cb27c451fbba2355580e99a0cdecca08e535b8917e4","last_reissued_at":"2026-07-05T05:27:22.098648Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:27:22.098648Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2212.10711","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":"2026-07-05T05:27:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lPZhBB6YEKfImHZ8z9u87xRP1ItgBGysVKvc1/x4mT+tdvhY/w3+U4skAGLWOxW3KbATZKgAyN4HlcxGb/0YBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T18:43:07.146276Z"},"content_sha256":"225cdd90cfea8071f142c0608991c136682e4b2419cf87fde86efdeee7a5bd1e","schema_version":"1.0","event_id":"sha256:225cdd90cfea8071f142c0608991c136682e4b2419cf87fde86efdeee7a5bd1e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:Y4MJW4VZPPQ3WV26HSZHYRI7XO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Task Ambiguity in Humans and Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Alex Tamkin, Avash Shrestha, Kunal Handa, Noah Goodman","submitted_at":"2022-12-20T18:35:33Z","abstract_excerpt":"Language models have recently achieved strong performance across a wide range of NLP benchmarks. However, unlike benchmarks, real world tasks are often poorly specified, and agents must deduce the user's intended behavior from a combination of context, instructions, and examples. We investigate how both humans and models behave in the face of such task ambiguity by proposing AmbiBench, a new benchmark of six ambiguously-specified classification tasks. We evaluate humans and models on AmbiBench by seeing how well they identify the intended task using 1) instructions with varying degrees of ambi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.10711","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/2212.10711/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":"2026-07-05T05:27:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7BZWnFq/UaFSVPqlgINg5z8vD6XNthj4hty28W5CIkh41J31RBtjynTHHnOTtZ2A6p6JXv1SLRXLnlQ+vgHGCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T18:43:07.147238Z"},"content_sha256":"6029959fcb8b258ef8c1abcbe6301600c1cfe8e5666ed29c9e6f1cc433f051f6","schema_version":"1.0","event_id":"sha256:6029959fcb8b258ef8c1abcbe6301600c1cfe8e5666ed29c9e6f1cc433f051f6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Y4MJW4VZPPQ3WV26HSZHYRI7XO/bundle.json","state_url":"https://pith.science/pith/Y4MJW4VZPPQ3WV26HSZHYRI7XO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Y4MJW4VZPPQ3WV26HSZHYRI7XO/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":"2026-08-14T18:43:07Z","links":{"resolver":"https://pith.science/pith/Y4MJW4VZPPQ3WV26HSZHYRI7XO","bundle":"https://pith.science/pith/Y4MJW4VZPPQ3WV26HSZHYRI7XO/bundle.json","state":"https://pith.science/pith/Y4MJW4VZPPQ3WV26HSZHYRI7XO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Y4MJW4VZPPQ3WV26HSZHYRI7XO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:Y4MJW4VZPPQ3WV26HSZHYRI7XO","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":"dbe322587788b5922a38d1c1a9253493885ec221b786f350bc070378eb5065ad","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-20T18:35:33Z","title_canon_sha256":"6a5f3abe1f0c1079a45f9c8eea70c5ccd6e7625987854dc924b66f72eecc5ee1"},"schema_version":"1.0","source":{"id":"2212.10711","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2212.10711","created_at":"2026-07-05T05:27:22Z"},{"alias_kind":"arxiv_version","alias_value":"2212.10711v1","created_at":"2026-07-05T05:27:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.10711","created_at":"2026-07-05T05:27:22Z"},{"alias_kind":"pith_short_12","alias_value":"Y4MJW4VZPPQ3","created_at":"2026-07-05T05:27:22Z"},{"alias_kind":"pith_short_16","alias_value":"Y4MJW4VZPPQ3WV26","created_at":"2026-07-05T05:27:22Z"},{"alias_kind":"pith_short_8","alias_value":"Y4MJW4VZ","created_at":"2026-07-05T05:27:22Z"}],"graph_snapshots":[{"event_id":"sha256:6029959fcb8b258ef8c1abcbe6301600c1cfe8e5666ed29c9e6f1cc433f051f6","target":"graph","created_at":"2026-07-05T05:27:22Z","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/2212.10711/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language models have recently achieved strong performance across a wide range of NLP benchmarks. However, unlike benchmarks, real world tasks are often poorly specified, and agents must deduce the user's intended behavior from a combination of context, instructions, and examples. We investigate how both humans and models behave in the face of such task ambiguity by proposing AmbiBench, a new benchmark of six ambiguously-specified classification tasks. We evaluate humans and models on AmbiBench by seeing how well they identify the intended task using 1) instructions with varying degrees of ambi","authors_text":"Alex Tamkin, Avash Shrestha, Kunal Handa, Noah Goodman","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-20T18:35:33Z","title":"Task Ambiguity in Humans and Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.10711","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":"sha256:225cdd90cfea8071f142c0608991c136682e4b2419cf87fde86efdeee7a5bd1e","target":"record","created_at":"2026-07-05T05:27:22Z","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":"dbe322587788b5922a38d1c1a9253493885ec221b786f350bc070378eb5065ad","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-20T18:35:33Z","title_canon_sha256":"6a5f3abe1f0c1079a45f9c8eea70c5ccd6e7625987854dc924b66f72eecc5ee1"},"schema_version":"1.0","source":{"id":"2212.10711","kind":"arxiv","version":1}},"canonical_sha256":"c7189b72b97be1bb575e3cb27c451fbba2355580e99a0cdecca08e535b8917e4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c7189b72b97be1bb575e3cb27c451fbba2355580e99a0cdecca08e535b8917e4","first_computed_at":"2026-07-05T05:27:22.098648Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:27:22.098648Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qmVxQOxt6Mpz/u9WMjPgPM+nVfEmuosOjRi0OkCcdMc+Fb3lMevGgWkvt+N4jijqR3ZL7cMlxnpwTf0EXoP/CA==","signature_status":"signed_v1","signed_at":"2026-07-05T05:27:22.099075Z","signed_message":"canonical_sha256_bytes"},"source_id":"2212.10711","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:225cdd90cfea8071f142c0608991c136682e4b2419cf87fde86efdeee7a5bd1e","sha256:6029959fcb8b258ef8c1abcbe6301600c1cfe8e5666ed29c9e6f1cc433f051f6"],"state_sha256":"d353159d7dbeb517128c3c064bff7070f16cb983a918ae739de0bc506db0adb7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8A9N6IhJNUbIkyxQvg6G7SoKvY3NJwMHd4EBO9wIiF55mtf7+Z51o8taWZLlsy51VSEdtNOyt3KU8Hh8W7qBCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T18:43:07.162812Z","bundle_sha256":"fce5b9b5cf479a551fd200f23bdf0095281897456c525d1b531c98f7023e6d3f"}}