{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:4UW7ZNCQXUAVEGQE3HRN22Z2GA","short_pith_number":"pith:4UW7ZNCQ","canonical_record":{"source":{"id":"2112.08321","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-12-15T18:10:54Z","cross_cats_sorted":[],"title_canon_sha256":"1ff428da36dc75307270a5ca32d3e80feff20fd980ca28c49e983cce7b00e578","abstract_canon_sha256":"4ffbd2f3a6010e779b39b89a40c307df85cd24e3731253557e1a188703ea0a16"},"schema_version":"1.0"},"canonical_sha256":"e52dfcb450bd01521a04d9e2dd6b3a302a94302a25b7fcba1e51cbdcbad13797","source":{"kind":"arxiv","id":"2112.08321","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2112.08321","created_at":"2026-07-05T05:13:05Z"},{"alias_kind":"arxiv_version","alias_value":"2112.08321v3","created_at":"2026-07-05T05:13:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.08321","created_at":"2026-07-05T05:13:05Z"},{"alias_kind":"pith_short_12","alias_value":"4UW7ZNCQXUAV","created_at":"2026-07-05T05:13:05Z"},{"alias_kind":"pith_short_16","alias_value":"4UW7ZNCQXUAVEGQE","created_at":"2026-07-05T05:13:05Z"},{"alias_kind":"pith_short_8","alias_value":"4UW7ZNCQ","created_at":"2026-07-05T05:13:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:4UW7ZNCQXUAVEGQE3HRN22Z2GA","target":"record","payload":{"canonical_record":{"source":{"id":"2112.08321","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-12-15T18:10:54Z","cross_cats_sorted":[],"title_canon_sha256":"1ff428da36dc75307270a5ca32d3e80feff20fd980ca28c49e983cce7b00e578","abstract_canon_sha256":"4ffbd2f3a6010e779b39b89a40c307df85cd24e3731253557e1a188703ea0a16"},"schema_version":"1.0"},"canonical_sha256":"e52dfcb450bd01521a04d9e2dd6b3a302a94302a25b7fcba1e51cbdcbad13797","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:13:05.013835Z","signature_b64":"rWTSJPRGttZXeOzc2L6FmL/iB64lJcLg+Ip74AEaWIR5pfJ6VhTi7INM7eRwqyrVk1W+HCluLjbZ/2Z/QdbLBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e52dfcb450bd01521a04d9e2dd6b3a302a94302a25b7fcba1e51cbdcbad13797","last_reissued_at":"2026-07-05T05:13:05.013390Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:13:05.013390Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2112.08321","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:13:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nkVfY8F99uPEmsXGmDrLQuHzLsqCdo517vSZbH4KJ5CsmFer8Av0WCZbmOZluqMu+w3SRIwTOPi+j8f+epi1Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T03:11:50.474602Z"},"content_sha256":"af1af714d3c9cd03b784232bd2ada654276813c85cade598185f4784b7a423f3","schema_version":"1.0","event_id":"sha256:af1af714d3c9cd03b784232bd2ada654276813c85cade598185f4784b7a423f3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:4UW7ZNCQXUAVEGQE3HRN22Z2GA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Know Thy Strengths: Comprehensive Dialogue State Tracking Diagnostics","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ahmad Beirami, Asli Celikyilmaz, Chinnadhurai Sankar, Christopher Lin, Hyundong Cho, Jonathan May, Kaushik Ram Sadagopan, Shahin Shayandeh","submitted_at":"2021-12-15T18:10:54Z","abstract_excerpt":"Recent works that revealed the vulnerability of dialogue state tracking (DST) models to distributional shifts have made holistic comparisons on robustness and qualitative analyses increasingly important for understanding their relative performance. We present our findings from standardized and comprehensive DST diagnoses, which have previously been sparse and uncoordinated, using our toolkit, CheckDST, a collection of robustness tests and failure mode analytics. We discover that different classes of DST models have clear strengths and weaknesses, where generation models are more promising for "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.08321","kind":"arxiv","version":3},"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/2112.08321/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":"2026-07-05T05:13:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"98fHxRrPlcm5a+FvJQCIj4yqCuv6OVAPZNKnCYupmOJylOvE3gRzGK/rPTt11oBCKzF7DzvEncbslcwxPXJLCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T03:11:50.479001Z"},"content_sha256":"c5ae0b82be5e137f15e2977d711b71c44b44a9fad5fe19ea61c9b00a5d661dce","schema_version":"1.0","event_id":"sha256:c5ae0b82be5e137f15e2977d711b71c44b44a9fad5fe19ea61c9b00a5d661dce"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4UW7ZNCQXUAVEGQE3HRN22Z2GA/bundle.json","state_url":"https://pith.science/pith/4UW7ZNCQXUAVEGQE3HRN22Z2GA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4UW7ZNCQXUAVEGQE3HRN22Z2GA/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":"2026-08-20T03:11:50Z","links":{"resolver":"https://pith.science/pith/4UW7ZNCQXUAVEGQE3HRN22Z2GA","bundle":"https://pith.science/pith/4UW7ZNCQXUAVEGQE3HRN22Z2GA/bundle.json","state":"https://pith.science/pith/4UW7ZNCQXUAVEGQE3HRN22Z2GA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4UW7ZNCQXUAVEGQE3HRN22Z2GA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:4UW7ZNCQXUAVEGQE3HRN22Z2GA","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":"4ffbd2f3a6010e779b39b89a40c307df85cd24e3731253557e1a188703ea0a16","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-12-15T18:10:54Z","title_canon_sha256":"1ff428da36dc75307270a5ca32d3e80feff20fd980ca28c49e983cce7b00e578"},"schema_version":"1.0","source":{"id":"2112.08321","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2112.08321","created_at":"2026-07-05T05:13:05Z"},{"alias_kind":"arxiv_version","alias_value":"2112.08321v3","created_at":"2026-07-05T05:13:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.08321","created_at":"2026-07-05T05:13:05Z"},{"alias_kind":"pith_short_12","alias_value":"4UW7ZNCQXUAV","created_at":"2026-07-05T05:13:05Z"},{"alias_kind":"pith_short_16","alias_value":"4UW7ZNCQXUAVEGQE","created_at":"2026-07-05T05:13:05Z"},{"alias_kind":"pith_short_8","alias_value":"4UW7ZNCQ","created_at":"2026-07-05T05:13:05Z"}],"graph_snapshots":[{"event_id":"sha256:c5ae0b82be5e137f15e2977d711b71c44b44a9fad5fe19ea61c9b00a5d661dce","target":"graph","created_at":"2026-07-05T05:13:05Z","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/2112.08321/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent works that revealed the vulnerability of dialogue state tracking (DST) models to distributional shifts have made holistic comparisons on robustness and qualitative analyses increasingly important for understanding their relative performance. We present our findings from standardized and comprehensive DST diagnoses, which have previously been sparse and uncoordinated, using our toolkit, CheckDST, a collection of robustness tests and failure mode analytics. We discover that different classes of DST models have clear strengths and weaknesses, where generation models are more promising for ","authors_text":"Ahmad Beirami, Asli Celikyilmaz, Chinnadhurai Sankar, Christopher Lin, Hyundong Cho, Jonathan May, Kaushik Ram Sadagopan, Shahin Shayandeh","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-12-15T18:10:54Z","title":"Know Thy Strengths: Comprehensive Dialogue State Tracking Diagnostics"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.08321","kind":"arxiv","version":3},"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:af1af714d3c9cd03b784232bd2ada654276813c85cade598185f4784b7a423f3","target":"record","created_at":"2026-07-05T05:13:05Z","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":"4ffbd2f3a6010e779b39b89a40c307df85cd24e3731253557e1a188703ea0a16","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-12-15T18:10:54Z","title_canon_sha256":"1ff428da36dc75307270a5ca32d3e80feff20fd980ca28c49e983cce7b00e578"},"schema_version":"1.0","source":{"id":"2112.08321","kind":"arxiv","version":3}},"canonical_sha256":"e52dfcb450bd01521a04d9e2dd6b3a302a94302a25b7fcba1e51cbdcbad13797","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e52dfcb450bd01521a04d9e2dd6b3a302a94302a25b7fcba1e51cbdcbad13797","first_computed_at":"2026-07-05T05:13:05.013390Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:13:05.013390Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rWTSJPRGttZXeOzc2L6FmL/iB64lJcLg+Ip74AEaWIR5pfJ6VhTi7INM7eRwqyrVk1W+HCluLjbZ/2Z/QdbLBg==","signature_status":"signed_v1","signed_at":"2026-07-05T05:13:05.013835Z","signed_message":"canonical_sha256_bytes"},"source_id":"2112.08321","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:af1af714d3c9cd03b784232bd2ada654276813c85cade598185f4784b7a423f3","sha256:c5ae0b82be5e137f15e2977d711b71c44b44a9fad5fe19ea61c9b00a5d661dce"],"state_sha256":"ff79890d2df219d22a874a45ccfaee08e58fffe598ec0fae6a81f8ef83d4d3db"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nQ+JF0pUMWq9WcrsTfn01aPcYjvLvTelgxBak8Opjx5uLRn7g0cjSuSrZfy3/yXhs836ZttCTsSEPF2cc8ASBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T03:11:50.486740Z","bundle_sha256":"cf02916506aa18e6006b05e975d12d24100c1c61067d33fd25a5b62b1b97153d"}}