{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:HUZWKCO2CPCN6NNY6ICQW7QS3P","short_pith_number":"pith:HUZWKCO2","canonical_record":{"source":{"id":"2505.24731","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T15:53:56Z","cross_cats_sorted":[],"title_canon_sha256":"e493752cef2596eb821e81479777ced0eaa71ed22ac81888645aa44e5ce40d76","abstract_canon_sha256":"13f06fe0e2b19bcc2a235dd69e2e5119f1f25984f77ed812b49d06bef89b05f6"},"schema_version":"1.0"},"canonical_sha256":"3d336509da13c4df35b8f2050b7e12dbeec879455ef59f29dddd5e1976978dd2","source":{"kind":"arxiv","id":"2505.24731","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.24731","created_at":"2026-07-05T11:12:55Z"},{"alias_kind":"arxiv_version","alias_value":"2505.24731v1","created_at":"2026-07-05T11:12:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24731","created_at":"2026-07-05T11:12:55Z"},{"alias_kind":"pith_short_12","alias_value":"HUZWKCO2CPCN","created_at":"2026-07-05T11:12:55Z"},{"alias_kind":"pith_short_16","alias_value":"HUZWKCO2CPCN6NNY","created_at":"2026-07-05T11:12:55Z"},{"alias_kind":"pith_short_8","alias_value":"HUZWKCO2","created_at":"2026-07-05T11:12:55Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:HUZWKCO2CPCN6NNY6ICQW7QS3P","target":"record","payload":{"canonical_record":{"source":{"id":"2505.24731","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T15:53:56Z","cross_cats_sorted":[],"title_canon_sha256":"e493752cef2596eb821e81479777ced0eaa71ed22ac81888645aa44e5ce40d76","abstract_canon_sha256":"13f06fe0e2b19bcc2a235dd69e2e5119f1f25984f77ed812b49d06bef89b05f6"},"schema_version":"1.0"},"canonical_sha256":"3d336509da13c4df35b8f2050b7e12dbeec879455ef59f29dddd5e1976978dd2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:55.706932Z","signature_b64":"wr74lutmKLjduH0kwVZwZmx0qHfZ2M3lKp6JD3my4LV2ZzCj5MRoXyFB9gZpUwkNuIZ40TgKRtmb2S5PzalyDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3d336509da13c4df35b8f2050b7e12dbeec879455ef59f29dddd5e1976978dd2","last_reissued_at":"2026-07-05T11:12:55.706422Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:55.706422Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.24731","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":"2026-07-05T11:12:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mak1Loyt1dqewrqaiVShqY/jCc89HW3GomsopsU8CMn4V4S06+89Qw3rX78uUBOUn+4NQWHAfrQ32xZxLiesDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T03:26:20.347869Z"},"content_sha256":"443126c33e3c0f2632d038a0a8557103beb5d2c8a0760eaa219a82f93c86efb1","schema_version":"1.0","event_id":"sha256:443126c33e3c0f2632d038a0a8557103beb5d2c8a0760eaa219a82f93c86efb1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:HUZWKCO2CPCN6NNY6ICQW7QS3P","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Circuit Stability Characterizes Language Model Generalization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alan Sun","submitted_at":"2025-05-30T15:53:56Z","abstract_excerpt":"Extensively evaluating the capabilities of (large) language models is difficult. Rapid development of state-of-the-art models induce benchmark saturation, while creating more challenging datasets is labor-intensive. Inspired by the recent developments in mechanistic interpretability, we introduce circuit stability as a new way to assess model performance. Circuit stability refers to a model's ability to apply a consistent reasoning process-its circuit-across various inputs. We mathematically formalize circuit stability and circuit equivalence. Then, through three case studies, we empirically s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24731","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/2505.24731/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":"2026-07-05T11:12:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"b7vDJ6wPJwFfm4iFrP4QWNGu+cqTVnTITY4A8StG/YHdhP2kysnR3/7e2o+koZbrTW49ZYPObje/VFlqq//hDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T03:26:20.348805Z"},"content_sha256":"fe26d66637aec2d5436bd3fd4015ad3902a07d23c416dfe9c3c9cbc9a8bc5a7b","schema_version":"1.0","event_id":"sha256:fe26d66637aec2d5436bd3fd4015ad3902a07d23c416dfe9c3c9cbc9a8bc5a7b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HUZWKCO2CPCN6NNY6ICQW7QS3P/bundle.json","state_url":"https://pith.science/pith/HUZWKCO2CPCN6NNY6ICQW7QS3P/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HUZWKCO2CPCN6NNY6ICQW7QS3P/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":"2026-08-10T03:26:20Z","links":{"resolver":"https://pith.science/pith/HUZWKCO2CPCN6NNY6ICQW7QS3P","bundle":"https://pith.science/pith/HUZWKCO2CPCN6NNY6ICQW7QS3P/bundle.json","state":"https://pith.science/pith/HUZWKCO2CPCN6NNY6ICQW7QS3P/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HUZWKCO2CPCN6NNY6ICQW7QS3P/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:HUZWKCO2CPCN6NNY6ICQW7QS3P","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":"13f06fe0e2b19bcc2a235dd69e2e5119f1f25984f77ed812b49d06bef89b05f6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T15:53:56Z","title_canon_sha256":"e493752cef2596eb821e81479777ced0eaa71ed22ac81888645aa44e5ce40d76"},"schema_version":"1.0","source":{"id":"2505.24731","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.24731","created_at":"2026-07-05T11:12:55Z"},{"alias_kind":"arxiv_version","alias_value":"2505.24731v1","created_at":"2026-07-05T11:12:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24731","created_at":"2026-07-05T11:12:55Z"},{"alias_kind":"pith_short_12","alias_value":"HUZWKCO2CPCN","created_at":"2026-07-05T11:12:55Z"},{"alias_kind":"pith_short_16","alias_value":"HUZWKCO2CPCN6NNY","created_at":"2026-07-05T11:12:55Z"},{"alias_kind":"pith_short_8","alias_value":"HUZWKCO2","created_at":"2026-07-05T11:12:55Z"}],"graph_snapshots":[{"event_id":"sha256:fe26d66637aec2d5436bd3fd4015ad3902a07d23c416dfe9c3c9cbc9a8bc5a7b","target":"graph","created_at":"2026-07-05T11:12:55Z","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/2505.24731/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Extensively evaluating the capabilities of (large) language models is difficult. Rapid development of state-of-the-art models induce benchmark saturation, while creating more challenging datasets is labor-intensive. Inspired by the recent developments in mechanistic interpretability, we introduce circuit stability as a new way to assess model performance. Circuit stability refers to a model's ability to apply a consistent reasoning process-its circuit-across various inputs. We mathematically formalize circuit stability and circuit equivalence. Then, through three case studies, we empirically s","authors_text":"Alan Sun","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T15:53:56Z","title":"Circuit Stability Characterizes Language Model Generalization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24731","kind":"arxiv","version":1},"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":"sha256:443126c33e3c0f2632d038a0a8557103beb5d2c8a0760eaa219a82f93c86efb1","target":"record","created_at":"2026-07-05T11:12:55Z","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":"13f06fe0e2b19bcc2a235dd69e2e5119f1f25984f77ed812b49d06bef89b05f6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T15:53:56Z","title_canon_sha256":"e493752cef2596eb821e81479777ced0eaa71ed22ac81888645aa44e5ce40d76"},"schema_version":"1.0","source":{"id":"2505.24731","kind":"arxiv","version":1}},"canonical_sha256":"3d336509da13c4df35b8f2050b7e12dbeec879455ef59f29dddd5e1976978dd2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3d336509da13c4df35b8f2050b7e12dbeec879455ef59f29dddd5e1976978dd2","first_computed_at":"2026-07-05T11:12:55.706422Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:12:55.706422Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wr74lutmKLjduH0kwVZwZmx0qHfZ2M3lKp6JD3my4LV2ZzCj5MRoXyFB9gZpUwkNuIZ40TgKRtmb2S5PzalyDg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:12:55.706932Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.24731","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:443126c33e3c0f2632d038a0a8557103beb5d2c8a0760eaa219a82f93c86efb1","sha256:fe26d66637aec2d5436bd3fd4015ad3902a07d23c416dfe9c3c9cbc9a8bc5a7b"],"state_sha256":"f2149201cc3264cd7a7d7b737766e9d24d24e7b6b4b327b377f0efe19da63587"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4rRv0nCFqHoOET13bzXGeI9ZTFex0fa+obvwHx5qL8XGSdKCNHd1M0/hDV1ZF5DdHUhU7TwrB/7tyJxpFTenAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T03:26:20.354892Z","bundle_sha256":"f992d20afab86a4a806f2dcb485a8261a21c67688fbb6422339b8f14f46a6632"}}