{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SQY5E4SUWHDJ7DHXVMFW3SOA3M","short_pith_number":"pith:SQY5E4SU","schema_version":"1.0","canonical_sha256":"9431d27254b1c69f8cf7ab0b6dc9c0db1862538b385b102eaf1ce3428b85d256","source":{"kind":"arxiv","id":"2606.06664","version":1},"attestation_state":"computed","paper":{"title":"Inside the Visual Mind: Neuroscience-Motivated Concept Circuits for Interpreting and Steering Vision Transformers","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Mengmeng Ma, Tang Li, Xi Peng, Yanlin Chen","submitted_at":"2026-06-04T19:25:40Z","abstract_excerpt":"Despite high accuracy, Vision Transformer (ViT) predictions can be driven by spurious cues, raising the need to understand their inner workings before safe deployment. Sparse autoencoders (SAEs) provide a promising lens for decomposing model representations into human-interpretable concepts, yet adapting SAE-based interpretation to ViTs remains challenging due to limited control over concept coverage and subjective, non-scalable feature interpretation. To fill the gaps, motivated by neuroscience-inspired principles, we propose ViSAE, a mechanistic interpretability toolbox for understanding ViT"},"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":"2606.06664","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-04T19:25:40Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"bc9b62d277801a445d2bb5433421dcbb889121f693469adcb35918b0b4361e12","abstract_canon_sha256":"d42784789d00beefc5e6ddcde38370383271ba0f274c75e21aeef0c88a3faf3a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-08T01:04:21.484255Z","signature_b64":"QP0SZ7TQUVQKoWM3H5cmFmbl2ptRfc08AcQhN44L4x7TFm2jQmjtkwLSlQ2oY0bj2/RyyTwab4i1zR4mVGC3AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9431d27254b1c69f8cf7ab0b6dc9c0db1862538b385b102eaf1ce3428b85d256","last_reissued_at":"2026-06-08T01:04:21.483491Z","signature_status":"signed_v1","first_computed_at":"2026-06-08T01:04:21.483491Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Inside the Visual Mind: Neuroscience-Motivated Concept Circuits for Interpreting and Steering Vision Transformers","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Mengmeng Ma, Tang Li, Xi Peng, Yanlin Chen","submitted_at":"2026-06-04T19:25:40Z","abstract_excerpt":"Despite high accuracy, Vision Transformer (ViT) predictions can be driven by spurious cues, raising the need to understand their inner workings before safe deployment. Sparse autoencoders (SAEs) provide a promising lens for decomposing model representations into human-interpretable concepts, yet adapting SAE-based interpretation to ViTs remains challenging due to limited control over concept coverage and subjective, non-scalable feature interpretation. To fill the gaps, motivated by neuroscience-inspired principles, we propose ViSAE, a mechanistic interpretability toolbox for understanding ViT"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.06664","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/2606.06664/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":"2606.06664","created_at":"2026-06-08T01:04:21.483624+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.06664v1","created_at":"2026-06-08T01:04:21.483624+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.06664","created_at":"2026-06-08T01:04:21.483624+00:00"},{"alias_kind":"pith_short_12","alias_value":"SQY5E4SUWHDJ","created_at":"2026-06-08T01:04:21.483624+00:00"},{"alias_kind":"pith_short_16","alias_value":"SQY5E4SUWHDJ7DHX","created_at":"2026-06-08T01:04:21.483624+00:00"},{"alias_kind":"pith_short_8","alias_value":"SQY5E4SU","created_at":"2026-06-08T01:04:21.483624+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SQY5E4SUWHDJ7DHXVMFW3SOA3M","json":"https://pith.science/pith/SQY5E4SUWHDJ7DHXVMFW3SOA3M.json","graph_json":"https://pith.science/api/pith-number/SQY5E4SUWHDJ7DHXVMFW3SOA3M/graph.json","events_json":"https://pith.science/api/pith-number/SQY5E4SUWHDJ7DHXVMFW3SOA3M/events.json","paper":"https://pith.science/paper/SQY5E4SU"},"agent_actions":{"view_html":"https://pith.science/pith/SQY5E4SUWHDJ7DHXVMFW3SOA3M","download_json":"https://pith.science/pith/SQY5E4SUWHDJ7DHXVMFW3SOA3M.json","view_paper":"https://pith.science/paper/SQY5E4SU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.06664&json=true","fetch_graph":"https://pith.science/api/pith-number/SQY5E4SUWHDJ7DHXVMFW3SOA3M/graph.json","fetch_events":"https://pith.science/api/pith-number/SQY5E4SUWHDJ7DHXVMFW3SOA3M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SQY5E4SUWHDJ7DHXVMFW3SOA3M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SQY5E4SUWHDJ7DHXVMFW3SOA3M/action/storage_attestation","attest_author":"https://pith.science/pith/SQY5E4SUWHDJ7DHXVMFW3SOA3M/action/author_attestation","sign_citation":"https://pith.science/pith/SQY5E4SUWHDJ7DHXVMFW3SOA3M/action/citation_signature","submit_replication":"https://pith.science/pith/SQY5E4SUWHDJ7DHXVMFW3SOA3M/action/replication_record"}},"created_at":"2026-06-08T01:04:21.483624+00:00","updated_at":"2026-06-08T01:04:21.483624+00:00"}