{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:5WTX6PWTBXLHKMRXOKAQYZQRDI","short_pith_number":"pith:5WTX6PWT","canonical_record":{"source":{"id":"2411.04430","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-07T04:52:18Z","cross_cats_sorted":[],"title_canon_sha256":"a798faa1c85f1fd8feb2db66681dd5069cc1db7fd3fa8a61fbf3aa2ca8d5e861","abstract_canon_sha256":"080aae073cc46f8c27f9b30a591e2924596b81c3626b1f801ed8baed1fdb4d59"},"schema_version":"1.0"},"canonical_sha256":"eda77f3ed30dd675323772810c66111a2fd6e960779f1140a90f6f1de1bfb501","source":{"kind":"arxiv","id":"2411.04430","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.04430","created_at":"2026-07-05T10:12:23Z"},{"alias_kind":"arxiv_version","alias_value":"2411.04430v2","created_at":"2026-07-05T10:12:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.04430","created_at":"2026-07-05T10:12:23Z"},{"alias_kind":"pith_short_12","alias_value":"5WTX6PWTBXLH","created_at":"2026-07-05T10:12:23Z"},{"alias_kind":"pith_short_16","alias_value":"5WTX6PWTBXLHKMRX","created_at":"2026-07-05T10:12:23Z"},{"alias_kind":"pith_short_8","alias_value":"5WTX6PWT","created_at":"2026-07-05T10:12:23Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:5WTX6PWTBXLHKMRXOKAQYZQRDI","target":"record","payload":{"canonical_record":{"source":{"id":"2411.04430","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-07T04:52:18Z","cross_cats_sorted":[],"title_canon_sha256":"a798faa1c85f1fd8feb2db66681dd5069cc1db7fd3fa8a61fbf3aa2ca8d5e861","abstract_canon_sha256":"080aae073cc46f8c27f9b30a591e2924596b81c3626b1f801ed8baed1fdb4d59"},"schema_version":"1.0"},"canonical_sha256":"eda77f3ed30dd675323772810c66111a2fd6e960779f1140a90f6f1de1bfb501","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:12:23.159517Z","signature_b64":"dw/+bWzQlVSw3Q+knNyAAgqKtDlnId+82o+Z/lTg9HURQGIbgzjYc/A41r22VlVeTEZXVz7UiaEhlZLw0cs7AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eda77f3ed30dd675323772810c66111a2fd6e960779f1140a90f6f1de1bfb501","last_reissued_at":"2026-07-05T10:12:23.159001Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:12:23.159001Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.04430","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:12:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qhMC5fLFi7e10+u/I7P5OYvayBv0EjhsBa6gAcjjoiv4UQpc90VtaKbx+gZno6gDpzGKPg20dBiPq77CnFvMAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T13:32:43.693882Z"},"content_sha256":"d4bf78494cd4e2c3e53652498e59c1bcce95c93f02367f4a11d555e084dc4d7a","schema_version":"1.0","event_id":"sha256:d4bf78494cd4e2c3e53652498e59c1bcce95c93f02367f4a11d555e084dc4d7a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:5WTX6PWTBXLHKMRXOKAQYZQRDI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Towards Unifying Interpretability and Control: Evaluation via Intervention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Asma Ghandeharioun, Himabindu Lakkaraju, Suraj Srinivas, Usha Bhalla","submitted_at":"2024-11-07T04:52:18Z","abstract_excerpt":"With the growing complexity and capability of large language models, a need to understand model reasoning has emerged, often motivated by an underlying goal of controlling and aligning models. While numerous interpretability and steering methods have been proposed as solutions, they are typically designed either for understanding or for control, seldom addressing both. Additionally, the lack of standardized applications, motivations, and evaluation metrics makes it difficult to assess methods' practical utility and efficacy. To address the aforementioned issues, we argue that intervention is a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.04430","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/2411.04430/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":"2026-07-05T10:12:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qizZ83DFl7QD9SkkXAGVRze4r0ZbqGkgKVpl7jhE7b5qYIRHuBp6aCelswEM3e7odA7+63voMOnPFkuatdrmCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T13:32:43.694626Z"},"content_sha256":"f9fb1adefe355482351ca59c2bf160cf7f06482323d41322d31a5fbaf8a0175e","schema_version":"1.0","event_id":"sha256:f9fb1adefe355482351ca59c2bf160cf7f06482323d41322d31a5fbaf8a0175e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5WTX6PWTBXLHKMRXOKAQYZQRDI/bundle.json","state_url":"https://pith.science/pith/5WTX6PWTBXLHKMRXOKAQYZQRDI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5WTX6PWTBXLHKMRXOKAQYZQRDI/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":"2026-08-21T13:32:43Z","links":{"resolver":"https://pith.science/pith/5WTX6PWTBXLHKMRXOKAQYZQRDI","bundle":"https://pith.science/pith/5WTX6PWTBXLHKMRXOKAQYZQRDI/bundle.json","state":"https://pith.science/pith/5WTX6PWTBXLHKMRXOKAQYZQRDI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5WTX6PWTBXLHKMRXOKAQYZQRDI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:5WTX6PWTBXLHKMRXOKAQYZQRDI","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":"080aae073cc46f8c27f9b30a591e2924596b81c3626b1f801ed8baed1fdb4d59","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-07T04:52:18Z","title_canon_sha256":"a798faa1c85f1fd8feb2db66681dd5069cc1db7fd3fa8a61fbf3aa2ca8d5e861"},"schema_version":"1.0","source":{"id":"2411.04430","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.04430","created_at":"2026-07-05T10:12:23Z"},{"alias_kind":"arxiv_version","alias_value":"2411.04430v2","created_at":"2026-07-05T10:12:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.04430","created_at":"2026-07-05T10:12:23Z"},{"alias_kind":"pith_short_12","alias_value":"5WTX6PWTBXLH","created_at":"2026-07-05T10:12:23Z"},{"alias_kind":"pith_short_16","alias_value":"5WTX6PWTBXLHKMRX","created_at":"2026-07-05T10:12:23Z"},{"alias_kind":"pith_short_8","alias_value":"5WTX6PWT","created_at":"2026-07-05T10:12:23Z"}],"graph_snapshots":[{"event_id":"sha256:f9fb1adefe355482351ca59c2bf160cf7f06482323d41322d31a5fbaf8a0175e","target":"graph","created_at":"2026-07-05T10:12:23Z","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/2411.04430/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With the growing complexity and capability of large language models, a need to understand model reasoning has emerged, often motivated by an underlying goal of controlling and aligning models. While numerous interpretability and steering methods have been proposed as solutions, they are typically designed either for understanding or for control, seldom addressing both. Additionally, the lack of standardized applications, motivations, and evaluation metrics makes it difficult to assess methods' practical utility and efficacy. To address the aforementioned issues, we argue that intervention is a","authors_text":"Asma Ghandeharioun, Himabindu Lakkaraju, Suraj Srinivas, Usha Bhalla","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-07T04:52:18Z","title":"Towards Unifying Interpretability and Control: Evaluation via Intervention"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.04430","kind":"arxiv","version":2},"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:d4bf78494cd4e2c3e53652498e59c1bcce95c93f02367f4a11d555e084dc4d7a","target":"record","created_at":"2026-07-05T10:12:23Z","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":"080aae073cc46f8c27f9b30a591e2924596b81c3626b1f801ed8baed1fdb4d59","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-07T04:52:18Z","title_canon_sha256":"a798faa1c85f1fd8feb2db66681dd5069cc1db7fd3fa8a61fbf3aa2ca8d5e861"},"schema_version":"1.0","source":{"id":"2411.04430","kind":"arxiv","version":2}},"canonical_sha256":"eda77f3ed30dd675323772810c66111a2fd6e960779f1140a90f6f1de1bfb501","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"eda77f3ed30dd675323772810c66111a2fd6e960779f1140a90f6f1de1bfb501","first_computed_at":"2026-07-05T10:12:23.159001Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:12:23.159001Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dw/+bWzQlVSw3Q+knNyAAgqKtDlnId+82o+Z/lTg9HURQGIbgzjYc/A41r22VlVeTEZXVz7UiaEhlZLw0cs7AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:12:23.159517Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.04430","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d4bf78494cd4e2c3e53652498e59c1bcce95c93f02367f4a11d555e084dc4d7a","sha256:f9fb1adefe355482351ca59c2bf160cf7f06482323d41322d31a5fbaf8a0175e"],"state_sha256":"5f1734f575bd80c74e31f80e85b798999df0e8beb6e131b3431c0254ef17d531"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"T/wFNwmXIvgDzuYqHaPQ0bVNBqzTW3iaqPPYCtpm1P9e/qljicm6Jxe9IMGJuGot4qyf4/B8AWAtr8TboYqxCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-21T13:32:43.699918Z","bundle_sha256":"eb6dd8eca14cfe7ec6e68c740f5ea5be050eb420bd0e53167e9e17085d1ac4c7"}}