{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZSBJZRHQ25NLWUZ6J7UDO3MW5F","short_pith_number":"pith:ZSBJZRHQ","schema_version":"1.0","canonical_sha256":"cc829cc4f0d75abb533e4fe8376d96e9704fede8e3358d827731633daa29872b","source":{"kind":"arxiv","id":"2606.30101","version":1},"attestation_state":"computed","paper":{"title":"SIR: Structured Image Representations for Explainable Robot Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Jan Schwab, Jens Bosch, Maximilian Xiling Li, Minh-Trung Tang, Moritz Haberland, Nils Blank, Paul Mattes, Rudolf Lioutikov","submitted_at":"2026-06-29T10:37:21Z","abstract_excerpt":"Existing robot policies based on learned visual embeddings lack explicit structure and are sensitive to visual distractions. Thus, the representations that drive their behaviour are often opaque, making their decision-making process difficult to interpret. To address this, we introduce Structured Image Representations (SIR), a method that leverages Scene Graphs (SGs) as an intermediate representation for robot policy learning. Our approach first constructs a fully connected graph, using image-derived features as initial node representations. Then, a module learns to sparsify this graph end-to-"},"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":"2606.30101","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-06-29T10:37:21Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"a1838a99b0562c0f9454147e9999bb699039266c5de4c5debb6cefddcbee2b3f","abstract_canon_sha256":"4562610c470bf81731c27947f5703eff8b4f905609ca9fd4e3f470de2fd981a6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T02:17:49.618139Z","signature_b64":"NHbL7pcKmGnC12uT3TCUUSJlSnNGs3CJhvBwsMH4mICMcf33ScDY20aekA9jtMLP7UWQD0wQUiGBr04GyuolDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cc829cc4f0d75abb533e4fe8376d96e9704fede8e3358d827731633daa29872b","last_reissued_at":"2026-06-30T02:17:49.617653Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T02:17:49.617653Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SIR: Structured Image Representations for Explainable Robot Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Jan Schwab, Jens Bosch, Maximilian Xiling Li, Minh-Trung Tang, Moritz Haberland, Nils Blank, Paul Mattes, Rudolf Lioutikov","submitted_at":"2026-06-29T10:37:21Z","abstract_excerpt":"Existing robot policies based on learned visual embeddings lack explicit structure and are sensitive to visual distractions. Thus, the representations that drive their behaviour are often opaque, making their decision-making process difficult to interpret. To address this, we introduce Structured Image Representations (SIR), a method that leverages Scene Graphs (SGs) as an intermediate representation for robot policy learning. Our approach first constructs a fully connected graph, using image-derived features as initial node representations. Then, a module learns to sparsify this graph end-to-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.30101","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/2606.30101/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":"2606.30101","created_at":"2026-06-30T02:17:49.617733+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.30101v1","created_at":"2026-06-30T02:17:49.617733+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.30101","created_at":"2026-06-30T02:17:49.617733+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZSBJZRHQ25NL","created_at":"2026-06-30T02:17:49.617733+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZSBJZRHQ25NLWUZ6","created_at":"2026-06-30T02:17:49.617733+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZSBJZRHQ","created_at":"2026-06-30T02:17:49.617733+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/ZSBJZRHQ25NLWUZ6J7UDO3MW5F","json":"https://pith.science/pith/ZSBJZRHQ25NLWUZ6J7UDO3MW5F.json","graph_json":"https://pith.science/api/pith-number/ZSBJZRHQ25NLWUZ6J7UDO3MW5F/graph.json","events_json":"https://pith.science/api/pith-number/ZSBJZRHQ25NLWUZ6J7UDO3MW5F/events.json","paper":"https://pith.science/paper/ZSBJZRHQ"},"agent_actions":{"view_html":"https://pith.science/pith/ZSBJZRHQ25NLWUZ6J7UDO3MW5F","download_json":"https://pith.science/pith/ZSBJZRHQ25NLWUZ6J7UDO3MW5F.json","view_paper":"https://pith.science/paper/ZSBJZRHQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.30101&json=true","fetch_graph":"https://pith.science/api/pith-number/ZSBJZRHQ25NLWUZ6J7UDO3MW5F/graph.json","fetch_events":"https://pith.science/api/pith-number/ZSBJZRHQ25NLWUZ6J7UDO3MW5F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZSBJZRHQ25NLWUZ6J7UDO3MW5F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZSBJZRHQ25NLWUZ6J7UDO3MW5F/action/storage_attestation","attest_author":"https://pith.science/pith/ZSBJZRHQ25NLWUZ6J7UDO3MW5F/action/author_attestation","sign_citation":"https://pith.science/pith/ZSBJZRHQ25NLWUZ6J7UDO3MW5F/action/citation_signature","submit_replication":"https://pith.science/pith/ZSBJZRHQ25NLWUZ6J7UDO3MW5F/action/replication_record"}},"created_at":"2026-06-30T02:17:49.617733+00:00","updated_at":"2026-06-30T02:17:49.617733+00:00"}