{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:P3NVYDKHAS4B6XJ5KJBBRFRKZS","short_pith_number":"pith:P3NVYDKH","schema_version":"1.0","canonical_sha256":"7edb5c0d4704b81f5d3d524218962acc9d18266b1235fa1f770ea12a98d17ca1","source":{"kind":"arxiv","id":"2412.03673","version":2},"attestation_state":"computed","paper":{"title":"Interpreting Transformers for Jet Tagging","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","hep-ex","physics.data-an"],"primary_cat":"hep-ph","authors_text":"Aaron Wang, Abhijith Gandrakota, Elham E Khoda, Javier Duarte, Jennifer Ngadiuba, Priyansh Bhatnagar, Vivekanand Sahu","submitted_at":"2024-12-04T19:06:40Z","abstract_excerpt":"Machine learning (ML) algorithms, particularly attention-based transformer models, have become indispensable for analyzing the vast data generated by particle physics experiments like ATLAS and CMS at the CERN LHC. Particle Transformer (ParT), a state-of-the-art model, leverages particle-level attention to improve jet-tagging tasks, which are critical for identifying particles resulting from proton collisions. This study focuses on interpreting ParT by analyzing attention heat maps and particle-pair correlations on the $\\eta$-$\\phi$ plane, revealing a binary attention pattern where each partic"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2412.03673","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"hep-ph","submitted_at":"2024-12-04T19:06:40Z","cross_cats_sorted":["cs.LG","hep-ex","physics.data-an"],"title_canon_sha256":"37aa9be3841e00dbfe9b8d8cb494e89a1681f89a301a9febb9c55a848b0efc1e","abstract_canon_sha256":"2c3e405ce383bf4ed5e59439d1d3861015fc1606b6ea93ad9bd9f7a8d31b66a3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:46:00.042389Z","signature_b64":"Ek7W5a4+DRYM97MenNqAGkZiWGIKJw+aNTd3Av981/kMqht6hFR0S4yfkXY0yNjHTwVr3SMwZVVq3nymF1AwDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7edb5c0d4704b81f5d3d524218962acc9d18266b1235fa1f770ea12a98d17ca1","last_reissued_at":"2026-07-05T09:46:00.041880Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:46:00.041880Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Interpreting Transformers for Jet Tagging","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","hep-ex","physics.data-an"],"primary_cat":"hep-ph","authors_text":"Aaron Wang, Abhijith Gandrakota, Elham E Khoda, Javier Duarte, Jennifer Ngadiuba, Priyansh Bhatnagar, Vivekanand Sahu","submitted_at":"2024-12-04T19:06:40Z","abstract_excerpt":"Machine learning (ML) algorithms, particularly attention-based transformer models, have become indispensable for analyzing the vast data generated by particle physics experiments like ATLAS and CMS at the CERN LHC. Particle Transformer (ParT), a state-of-the-art model, leverages particle-level attention to improve jet-tagging tasks, which are critical for identifying particles resulting from proton collisions. This study focuses on interpreting ParT by analyzing attention heat maps and particle-pair correlations on the $\\eta$-$\\phi$ plane, revealing a binary attention pattern where each partic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.03673","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/2412.03673/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2412.03673","created_at":"2026-07-05T09:46:00.041945+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.03673v2","created_at":"2026-07-05T09:46:00.041945+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.03673","created_at":"2026-07-05T09:46:00.041945+00:00"},{"alias_kind":"pith_short_12","alias_value":"P3NVYDKHAS4B","created_at":"2026-07-05T09:46:00.041945+00:00"},{"alias_kind":"pith_short_16","alias_value":"P3NVYDKHAS4B6XJ5","created_at":"2026-07-05T09:46:00.041945+00:00"},{"alias_kind":"pith_short_8","alias_value":"P3NVYDKH","created_at":"2026-07-05T09:46:00.041945+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21790","citing_title":"What Do Lorentz-Equivariant Jet Taggers Learn?","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2505.21042","citing_title":"What exactly did the Transformer learn from our physics data?","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2508.15468","citing_title":"JEDI-linear: Fast and Efficient Graph Neural Networks for Jet Tagging on FPGAs","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09881","citing_title":"Dissecting Jet-Tagger Through Mechanistic Interpretability","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25885","citing_title":"Explainable AI for Jet Tagging: A Comparative Study of GNNExplainer, GNNShap, and GradCAM for Jet Tagging in the Lund Jet Plane","ref_index":44,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P3NVYDKHAS4B6XJ5KJBBRFRKZS","json":"https://pith.science/pith/P3NVYDKHAS4B6XJ5KJBBRFRKZS.json","graph_json":"https://pith.science/api/pith-number/P3NVYDKHAS4B6XJ5KJBBRFRKZS/graph.json","events_json":"https://pith.science/api/pith-number/P3NVYDKHAS4B6XJ5KJBBRFRKZS/events.json","paper":"https://pith.science/paper/P3NVYDKH"},"agent_actions":{"view_html":"https://pith.science/pith/P3NVYDKHAS4B6XJ5KJBBRFRKZS","download_json":"https://pith.science/pith/P3NVYDKHAS4B6XJ5KJBBRFRKZS.json","view_paper":"https://pith.science/paper/P3NVYDKH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.03673&json=true","fetch_graph":"https://pith.science/api/pith-number/P3NVYDKHAS4B6XJ5KJBBRFRKZS/graph.json","fetch_events":"https://pith.science/api/pith-number/P3NVYDKHAS4B6XJ5KJBBRFRKZS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P3NVYDKHAS4B6XJ5KJBBRFRKZS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P3NVYDKHAS4B6XJ5KJBBRFRKZS/action/storage_attestation","attest_author":"https://pith.science/pith/P3NVYDKHAS4B6XJ5KJBBRFRKZS/action/author_attestation","sign_citation":"https://pith.science/pith/P3NVYDKHAS4B6XJ5KJBBRFRKZS/action/citation_signature","submit_replication":"https://pith.science/pith/P3NVYDKHAS4B6XJ5KJBBRFRKZS/action/replication_record"}},"created_at":"2026-07-05T09:46:00.041945+00:00","updated_at":"2026-07-05T09:46:00.041945+00:00"}