{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:JUGEBUZRR6DH553GPJVZDBVW2N","short_pith_number":"pith:JUGEBUZR","canonical_record":{"source":{"id":"2605.13830","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T17:52:19Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7cdca6a47c21d818a73a33d19a1e6c3cc3394e85214f076cea826e887fa3a76d","abstract_canon_sha256":"1685f66660259bde9ccaaddc866196073e35e37331d821d4210b695b65937b5d"},"schema_version":"1.0"},"canonical_sha256":"4d0c40d3318f867ef7667a6b9186b6d3403f017414d064cab9ca56a2ab6f1ba2","source":{"kind":"arxiv","id":"2605.13830","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13830","created_at":"2026-05-18T02:44:15Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13830v1","created_at":"2026-05-18T02:44:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13830","created_at":"2026-05-18T02:44:15Z"},{"alias_kind":"pith_short_12","alias_value":"JUGEBUZRR6DH","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"JUGEBUZRR6DH553G","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"JUGEBUZR","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:JUGEBUZRR6DH553GPJVZDBVW2N","target":"record","payload":{"canonical_record":{"source":{"id":"2605.13830","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T17:52:19Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7cdca6a47c21d818a73a33d19a1e6c3cc3394e85214f076cea826e887fa3a76d","abstract_canon_sha256":"1685f66660259bde9ccaaddc866196073e35e37331d821d4210b695b65937b5d"},"schema_version":"1.0"},"canonical_sha256":"4d0c40d3318f867ef7667a6b9186b6d3403f017414d064cab9ca56a2ab6f1ba2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:15.074369Z","signature_b64":"Ii03VarAtikykZSEeFinMXyz9Q1KZJGqNGIwYB4quCTE8XJwtIFZ14c77DGmKzvI/So5vIeLYTc9oWf85ozhDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4d0c40d3318f867ef7667a6b9186b6d3403f017414d064cab9ca56a2ab6f1ba2","last_reissued_at":"2026-05-18T02:44:15.073786Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:15.073786Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.13830","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":"2026-05-18T02:44:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tO8bQF7FGshB+MFcVduqLcAEqZIN3bw9QmiZYRVSRbfv0EYrm21EXZ4866q569JrVn8fQlqgdrdM69ihxQDzDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T02:32:26.787935Z"},"content_sha256":"0aa4351fb00873f6574b33134c4756170e9e6cf88cfdd465b9d8b40ea9d7f04d","schema_version":"1.0","event_id":"sha256:0aa4351fb00873f6574b33134c4756170e9e6cf88cfdd465b9d8b40ea9d7f04d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:JUGEBUZRR6DH553GPJVZDBVW2N","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Quantifying Sensitivity for Tree Ensembles: A symbolic and compositional approach","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Decision tree ensembles can have their sensitivity to small input changes quantified by discretizing the space and counting susceptible regions via algebraic decision diagrams.","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Ajinkya Naik, Ashutosh Gupta, Chaitanya Garg, Kuldeep S. Meel, S. Akshay","submitted_at":"2026-05-13T17:52:19Z","abstract_excerpt":"Decision tree ensembles (DTE) are a popular model for a wide range of AI classification tasks, used in multiple safety critical domains, and hence verifying properties on these models has been an active topic of study over the last decade. One such verification question is the problem of sensitivity, which asks, given a DTE, whether a small change in subset of features can lead to misclassification of the input. In this work, our focus is to build a quantitative notion of sensitivity, tailored to DTEs, by discretizing the input space of the model and enumerating the regions which are susceptib"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We propose a novel algorithmic technique that can perform this computation efficiently, within a certified error and confidence bound. Our approach is based on encoding the problem as an algebraic decision diagram (ADD), and further splitting it into subproblems that can be solved efficiently and make the computation compositional and scalable.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The discretization of the input space combined with the ADD encoding accurately captures all sensitivity regions without introducing unaccounted approximation errors that affect the certified bounds.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A compositional algebraic decision diagram algorithm quantifies sensitivity in decision tree ensembles with certified error and confidence bounds, outperforming model counters on benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Decision tree ensembles can have their sensitivity to small input changes quantified by discretizing the space and counting susceptible regions via algebraic decision diagrams.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"c44dd1562f0c206e0186f0236c01f054c518e44ccfdf1733ca48cd5b33a9e6ca"},"source":{"id":"2605.13830","kind":"arxiv","version":1},"verdict":{"id":"4a7794a4-8211-4e32-a944-eca969f56146","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T17:45:33.735570Z","strongest_claim":"We propose a novel algorithmic technique that can perform this computation efficiently, within a certified error and confidence bound. Our approach is based on encoding the problem as an algebraic decision diagram (ADD), and further splitting it into subproblems that can be solved efficiently and make the computation compositional and scalable.","one_line_summary":"A compositional algebraic decision diagram algorithm quantifies sensitivity in decision tree ensembles with certified error and confidence bounds, outperforming model counters on benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The discretization of the input space combined with the ADD encoding accurately captures all sensitivity regions without introducing unaccounted approximation errors that affect the certified bounds.","pith_extraction_headline":"Decision tree ensembles can have their sensitivity to small input changes quantified by discretizing the space and counting susceptible regions via algebraic decision diagrams."},"references":{"count":47,"sample":[{"doi":"","year":2025,"title":"In: The Thirteenth International Conference on Learning Representations,ICLR2025,Singapore,April24-28,2025.OpenReview.net(2025), https://openreview.net/forum?id=h0vC0fm1q7","work_id":"2f24e1cd-d6d4-4df2-aeb3-883a2ad8c957","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1007/978-3-031-65630-9","year":2024,"title":"In: Gurfinkel, A., Ganesh, V","work_id":"14e6132a-5744-4012-a15e-8c4317785144","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1997,"title":"Formal methods in system design10(2), 171–206 (1997)","work_id":"0d326bdc-a77d-4d6b-a883-1ab8c4cb2ed3","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1109/icse48619.2023.00134","year":2023,"title":"In2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE)","work_id":"5955f330-cfab-45b0-9ff5-4dd04bca728a","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.3389/frai.2023.1124553","year":2023,"title":"Journal of Computational Physics493, 112455 (2023) https://doi.org/10.1016/j.jcp.2023.112455","work_id":"9d72a7e0-107b-45ae-a7b3-adbf9657b8e6","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":47,"snapshot_sha256":"49c796f71a2e044288d1d768bfa4f6c3645612553098d99201d767ea5e690467","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":"4a7794a4-8211-4e32-a944-eca969f56146"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T02:44:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YqrmlyShGwjzHTzsOTU7E82tR48H++E5ltASr5I5MYi27ktEjhLxjgOA70R8GBTCvf9Ia4QOINQ1GuhE+UNQCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T02:32:26.789133Z"},"content_sha256":"d0db1be42574cfb1fb8f5b970eafef8d8c68359eb1a3bcf4a19093655a899476","schema_version":"1.0","event_id":"sha256:d0db1be42574cfb1fb8f5b970eafef8d8c68359eb1a3bcf4a19093655a899476"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JUGEBUZRR6DH553GPJVZDBVW2N/bundle.json","state_url":"https://pith.science/pith/JUGEBUZRR6DH553GPJVZDBVW2N/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JUGEBUZRR6DH553GPJVZDBVW2N/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":"2026-05-23T02:32:26Z","links":{"resolver":"https://pith.science/pith/JUGEBUZRR6DH553GPJVZDBVW2N","bundle":"https://pith.science/pith/JUGEBUZRR6DH553GPJVZDBVW2N/bundle.json","state":"https://pith.science/pith/JUGEBUZRR6DH553GPJVZDBVW2N/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JUGEBUZRR6DH553GPJVZDBVW2N/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:JUGEBUZRR6DH553GPJVZDBVW2N","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":"1685f66660259bde9ccaaddc866196073e35e37331d821d4210b695b65937b5d","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T17:52:19Z","title_canon_sha256":"7cdca6a47c21d818a73a33d19a1e6c3cc3394e85214f076cea826e887fa3a76d"},"schema_version":"1.0","source":{"id":"2605.13830","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13830","created_at":"2026-05-18T02:44:15Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13830v1","created_at":"2026-05-18T02:44:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13830","created_at":"2026-05-18T02:44:15Z"},{"alias_kind":"pith_short_12","alias_value":"JUGEBUZRR6DH","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"JUGEBUZRR6DH553G","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"JUGEBUZR","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:d0db1be42574cfb1fb8f5b970eafef8d8c68359eb1a3bcf4a19093655a899476","target":"graph","created_at":"2026-05-18T02:44:15Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"We propose a novel algorithmic technique that can perform this computation efficiently, within a certified error and confidence bound. Our approach is based on encoding the problem as an algebraic decision diagram (ADD), and further splitting it into subproblems that can be solved efficiently and make the computation compositional and scalable."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The discretization of the input space combined with the ADD encoding accurately captures all sensitivity regions without introducing unaccounted approximation errors that affect the certified bounds."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A compositional algebraic decision diagram algorithm quantifies sensitivity in decision tree ensembles with certified error and confidence bounds, outperforming model counters on benchmarks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Decision tree ensembles can have their sensitivity to small input changes quantified by discretizing the space and counting susceptible regions via algebraic decision diagrams."}],"snapshot_sha256":"c44dd1562f0c206e0186f0236c01f054c518e44ccfdf1733ca48cd5b33a9e6ca"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Decision tree ensembles (DTE) are a popular model for a wide range of AI classification tasks, used in multiple safety critical domains, and hence verifying properties on these models has been an active topic of study over the last decade. One such verification question is the problem of sensitivity, which asks, given a DTE, whether a small change in subset of features can lead to misclassification of the input. In this work, our focus is to build a quantitative notion of sensitivity, tailored to DTEs, by discretizing the input space of the model and enumerating the regions which are susceptib","authors_text":"Ajinkya Naik, Ashutosh Gupta, Chaitanya Garg, Kuldeep S. Meel, S. Akshay","cross_cats":["cs.LG"],"headline":"Decision tree ensembles can have their sensitivity to small input changes quantified by discretizing the space and counting susceptible regions via algebraic decision diagrams.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T17:52:19Z","title":"Quantifying Sensitivity for Tree Ensembles: A symbolic and compositional approach"},"references":{"count":47,"internal_anchors":0,"resolved_work":47,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"In: The Thirteenth International Conference on Learning Representations,ICLR2025,Singapore,April24-28,2025.OpenReview.net(2025), https://openreview.net/forum?id=h0vC0fm1q7","work_id":"2f24e1cd-d6d4-4df2-aeb3-883a2ad8c957","year":2025},{"cited_arxiv_id":"","doi":"10.1007/978-3-031-65630-9","is_internal_anchor":false,"ref_index":2,"title":"In: Gurfinkel, A., Ganesh, V","work_id":"14e6132a-5744-4012-a15e-8c4317785144","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Formal methods in system design10(2), 171–206 (1997)","work_id":"0d326bdc-a77d-4d6b-a883-1ab8c4cb2ed3","year":1997},{"cited_arxiv_id":"","doi":"10.1109/icse48619.2023.00134","is_internal_anchor":false,"ref_index":4,"title":"In2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE)","work_id":"5955f330-cfab-45b0-9ff5-4dd04bca728a","year":2023},{"cited_arxiv_id":"","doi":"10.3389/frai.2023.1124553","is_internal_anchor":false,"ref_index":5,"title":"Journal of Computational Physics493, 112455 (2023) https://doi.org/10.1016/j.jcp.2023.112455","work_id":"9d72a7e0-107b-45ae-a7b3-adbf9657b8e6","year":2023}],"snapshot_sha256":"49c796f71a2e044288d1d768bfa4f6c3645612553098d99201d767ea5e690467"},"source":{"id":"2605.13830","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T17:45:33.735570Z","id":"4a7794a4-8211-4e32-a944-eca969f56146","model_set":{"reader":"grok-4.3"},"one_line_summary":"A compositional algebraic decision diagram algorithm quantifies sensitivity in decision tree ensembles with certified error and confidence bounds, outperforming model counters on benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Decision tree ensembles can have their sensitivity to small input changes quantified by discretizing the space and counting susceptible regions via algebraic decision diagrams.","strongest_claim":"We propose a novel algorithmic technique that can perform this computation efficiently, within a certified error and confidence bound. Our approach is based on encoding the problem as an algebraic decision diagram (ADD), and further splitting it into subproblems that can be solved efficiently and make the computation compositional and scalable.","weakest_assumption":"The discretization of the input space combined with the ADD encoding accurately captures all sensitivity regions without introducing unaccounted approximation errors that affect the certified bounds."}},"verdict_id":"4a7794a4-8211-4e32-a944-eca969f56146"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0aa4351fb00873f6574b33134c4756170e9e6cf88cfdd465b9d8b40ea9d7f04d","target":"record","created_at":"2026-05-18T02:44:15Z","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":"1685f66660259bde9ccaaddc866196073e35e37331d821d4210b695b65937b5d","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T17:52:19Z","title_canon_sha256":"7cdca6a47c21d818a73a33d19a1e6c3cc3394e85214f076cea826e887fa3a76d"},"schema_version":"1.0","source":{"id":"2605.13830","kind":"arxiv","version":1}},"canonical_sha256":"4d0c40d3318f867ef7667a6b9186b6d3403f017414d064cab9ca56a2ab6f1ba2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4d0c40d3318f867ef7667a6b9186b6d3403f017414d064cab9ca56a2ab6f1ba2","first_computed_at":"2026-05-18T02:44:15.073786Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T02:44:15.073786Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Ii03VarAtikykZSEeFinMXyz9Q1KZJGqNGIwYB4quCTE8XJwtIFZ14c77DGmKzvI/So5vIeLYTc9oWf85ozhDA==","signature_status":"signed_v1","signed_at":"2026-05-18T02:44:15.074369Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.13830","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0aa4351fb00873f6574b33134c4756170e9e6cf88cfdd465b9d8b40ea9d7f04d","sha256:d0db1be42574cfb1fb8f5b970eafef8d8c68359eb1a3bcf4a19093655a899476"],"state_sha256":"4b3b23ce00cc5569aa7186b1f20aee480088261dd8db96961da7ef7cdd7b55df"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"R1btX6+mBL6f9B6z4BNpXnSEO7wrz1+bGofmkxrw2B05dq6orje1aB4g9WQDX3YJFRZamzzJpOa9RaoOqFFFBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-23T02:32:26.794996Z","bundle_sha256":"97851b925457825e8d005371baf9fad92eff29e7a10dbab9bb1b150d1384d32c"}}