{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:6QJ4YT2FKBD3EZLMMC62IAYLFG","short_pith_number":"pith:6QJ4YT2F","schema_version":"1.0","canonical_sha256":"f413cc4f455047b2656c60bda4030b298a76faddc9c92f235828f853aebd2c47","source":{"kind":"arxiv","id":"2403.07809","version":1},"attestation_state":"computed","paper":{"title":"pyvene: A Library for Understanding and Improving PyTorch Models via Interventions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Aryaman Arora, Atticus Geiger, Christopher D. Manning, Christopher Potts, Jing Huang, Noah D. Goodman, Zheng Wang, Zhengxuan Wu","submitted_at":"2024-03-12T16:46:54Z","abstract_excerpt":"Interventions on model-internal states are fundamental operations in many areas of AI, including model editing, steering, robustness, and interpretability. To facilitate such research, we introduce $\\textbf{pyvene}$, an open-source Python library that supports customizable interventions on a range of different PyTorch modules. $\\textbf{pyvene}$ supports complex intervention schemes with an intuitive configuration format, and its interventions can be static or include trainable parameters. We show how $\\textbf{pyvene}$ provides a unified and extensible framework for performing interventions on "},"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":"2403.07809","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-12T16:46:54Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"f6130a577ec0bbb5313cd199310f7f5369175f5c78db8381928a24ea037e5e6b","abstract_canon_sha256":"ae2c75ad51299a4b7df6501c4b43702e83175fabea3a64f47edf6ec28f80c39a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:55:15.755391Z","signature_b64":"IsSxaD+g6pVIpx76szZK6W/1P4KNr3iIRfL3IkmjJ/e+a+LLMwylCtrog723rHwNmmmbzKqZnwd1DPOPdffrCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f413cc4f455047b2656c60bda4030b298a76faddc9c92f235828f853aebd2c47","last_reissued_at":"2026-07-05T07:55:15.754872Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:55:15.754872Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"pyvene: A Library for Understanding and Improving PyTorch Models via Interventions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Aryaman Arora, Atticus Geiger, Christopher D. Manning, Christopher Potts, Jing Huang, Noah D. Goodman, Zheng Wang, Zhengxuan Wu","submitted_at":"2024-03-12T16:46:54Z","abstract_excerpt":"Interventions on model-internal states are fundamental operations in many areas of AI, including model editing, steering, robustness, and interpretability. To facilitate such research, we introduce $\\textbf{pyvene}$, an open-source Python library that supports customizable interventions on a range of different PyTorch modules. $\\textbf{pyvene}$ supports complex intervention schemes with an intuitive configuration format, and its interventions can be static or include trainable parameters. We show how $\\textbf{pyvene}$ provides a unified and extensible framework for performing interventions on "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.07809","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/2403.07809/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":"2403.07809","created_at":"2026-07-05T07:55:15.754935+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.07809v1","created_at":"2026-07-05T07:55:15.754935+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.07809","created_at":"2026-07-05T07:55:15.754935+00:00"},{"alias_kind":"pith_short_12","alias_value":"6QJ4YT2FKBD3","created_at":"2026-07-05T07:55:15.754935+00:00"},{"alias_kind":"pith_short_16","alias_value":"6QJ4YT2FKBD3EZLM","created_at":"2026-07-05T07:55:15.754935+00:00"},{"alias_kind":"pith_short_8","alias_value":"6QJ4YT2F","created_at":"2026-07-05T07:55:15.754935+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19594","citing_title":"Unsupervised Causal Abstractions Discovery","ref_index":90,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02234","citing_title":"Bucketing the Good Apples: A Method for Diagnosing and Improving Causal Abstraction","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6QJ4YT2FKBD3EZLMMC62IAYLFG","json":"https://pith.science/pith/6QJ4YT2FKBD3EZLMMC62IAYLFG.json","graph_json":"https://pith.science/api/pith-number/6QJ4YT2FKBD3EZLMMC62IAYLFG/graph.json","events_json":"https://pith.science/api/pith-number/6QJ4YT2FKBD3EZLMMC62IAYLFG/events.json","paper":"https://pith.science/paper/6QJ4YT2F"},"agent_actions":{"view_html":"https://pith.science/pith/6QJ4YT2FKBD3EZLMMC62IAYLFG","download_json":"https://pith.science/pith/6QJ4YT2FKBD3EZLMMC62IAYLFG.json","view_paper":"https://pith.science/paper/6QJ4YT2F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.07809&json=true","fetch_graph":"https://pith.science/api/pith-number/6QJ4YT2FKBD3EZLMMC62IAYLFG/graph.json","fetch_events":"https://pith.science/api/pith-number/6QJ4YT2FKBD3EZLMMC62IAYLFG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6QJ4YT2FKBD3EZLMMC62IAYLFG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6QJ4YT2FKBD3EZLMMC62IAYLFG/action/storage_attestation","attest_author":"https://pith.science/pith/6QJ4YT2FKBD3EZLMMC62IAYLFG/action/author_attestation","sign_citation":"https://pith.science/pith/6QJ4YT2FKBD3EZLMMC62IAYLFG/action/citation_signature","submit_replication":"https://pith.science/pith/6QJ4YT2FKBD3EZLMMC62IAYLFG/action/replication_record"}},"created_at":"2026-07-05T07:55:15.754935+00:00","updated_at":"2026-07-05T07:55:15.754935+00:00"}