{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FNUFVVQNVVYNKPNJRII2RXFUPK","short_pith_number":"pith:FNUFVVQN","schema_version":"1.0","canonical_sha256":"2b685ad60dad70d53da98a11a8dcb47a83a52c00f50eaa0a87a72d7ca33c1d34","source":{"kind":"arxiv","id":"2605.17204","version":1},"attestation_state":"computed","paper":{"title":"Event-Grounded Sparse Autoencoders for Vision-Language-Action Policies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Aditya Chatterjee, Pranav Kumar, Rohan Paleja, Xinchen Jin","submitted_at":"2026-05-17T00:20:17Z","abstract_excerpt":"Vision-Language-Action (VLA) policies translate language and visual inputs into robot actions, where their hidden representations directly shape closed-loop behavior. However, mechanistic interpretability tools from language and vision-language models do not transfer cleanly to VLAs: outputs are robot actions rather than human-readable tokens, and interventions can only be tested via expensive closed-loop rollouts. We propose an event-grounded interpretability pipeline that anchors SAE feature analysis to behavioral events rather than text contexts. End-effector keyframes are clustered within "},"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":"2605.17204","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-05-17T00:20:17Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8968457de2afd1e9b4f6bc58cc0798edb61a06481b5af52f3a789c26b7dad2a6","abstract_canon_sha256":"cd082e026d5936a3df24f78354348a2beda7348333d670de2005c954eeb2a8ab"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:03:45.015391Z","signature_b64":"mzlXCUN5a4rJ1Q2R1/aGc6MHaUE6a8mfxidzNX0YbqU8lFAYKKHJdL8wrXB1tWlx4z9jfdQ7VjtuyK5qqoX9BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2b685ad60dad70d53da98a11a8dcb47a83a52c00f50eaa0a87a72d7ca33c1d34","last_reissued_at":"2026-05-20T00:03:45.014574Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:03:45.014574Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Event-Grounded Sparse Autoencoders for Vision-Language-Action Policies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Aditya Chatterjee, Pranav Kumar, Rohan Paleja, Xinchen Jin","submitted_at":"2026-05-17T00:20:17Z","abstract_excerpt":"Vision-Language-Action (VLA) policies translate language and visual inputs into robot actions, where their hidden representations directly shape closed-loop behavior. However, mechanistic interpretability tools from language and vision-language models do not transfer cleanly to VLAs: outputs are robot actions rather than human-readable tokens, and interventions can only be tested via expensive closed-loop rollouts. We propose an event-grounded interpretability pipeline that anchors SAE feature analysis to behavioral events rather than text contexts. End-effector keyframes are clustered within "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.17204","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/2605.17204/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T22:33:23.730667Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T22:01:57.939706Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"4c64fcb016f84f84b9d8c73a0f716b20596b29e8660c68d648b3d367a0142a8a"},"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":"2605.17204","created_at":"2026-05-20T00:03:45.014686+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.17204v1","created_at":"2026-05-20T00:03:45.014686+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17204","created_at":"2026-05-20T00:03:45.014686+00:00"},{"alias_kind":"pith_short_12","alias_value":"FNUFVVQNVVYN","created_at":"2026-05-20T00:03:45.014686+00:00"},{"alias_kind":"pith_short_16","alias_value":"FNUFVVQNVVYNKPNJ","created_at":"2026-05-20T00:03:45.014686+00:00"},{"alias_kind":"pith_short_8","alias_value":"FNUFVVQN","created_at":"2026-05-20T00:03:45.014686+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/FNUFVVQNVVYNKPNJRII2RXFUPK","json":"https://pith.science/pith/FNUFVVQNVVYNKPNJRII2RXFUPK.json","graph_json":"https://pith.science/api/pith-number/FNUFVVQNVVYNKPNJRII2RXFUPK/graph.json","events_json":"https://pith.science/api/pith-number/FNUFVVQNVVYNKPNJRII2RXFUPK/events.json","paper":"https://pith.science/paper/FNUFVVQN"},"agent_actions":{"view_html":"https://pith.science/pith/FNUFVVQNVVYNKPNJRII2RXFUPK","download_json":"https://pith.science/pith/FNUFVVQNVVYNKPNJRII2RXFUPK.json","view_paper":"https://pith.science/paper/FNUFVVQN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.17204&json=true","fetch_graph":"https://pith.science/api/pith-number/FNUFVVQNVVYNKPNJRII2RXFUPK/graph.json","fetch_events":"https://pith.science/api/pith-number/FNUFVVQNVVYNKPNJRII2RXFUPK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FNUFVVQNVVYNKPNJRII2RXFUPK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FNUFVVQNVVYNKPNJRII2RXFUPK/action/storage_attestation","attest_author":"https://pith.science/pith/FNUFVVQNVVYNKPNJRII2RXFUPK/action/author_attestation","sign_citation":"https://pith.science/pith/FNUFVVQNVVYNKPNJRII2RXFUPK/action/citation_signature","submit_replication":"https://pith.science/pith/FNUFVVQNVVYNKPNJRII2RXFUPK/action/replication_record"}},"created_at":"2026-05-20T00:03:45.014686+00:00","updated_at":"2026-05-20T00:03:45.014686+00:00"}