{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ZUBHSWYDNJK3MBB5CJZUEKGSC4","short_pith_number":"pith:ZUBHSWYD","schema_version":"1.0","canonical_sha256":"cd02795b036a55b6043d12734228d217348e3cf75c4f769330f656e05f4e8a81","source":{"kind":"arxiv","id":"2311.16680","version":2},"attestation_state":"computed","paper":{"title":"ROSO: Improving Robotic Policy Inference via Synthetic Observations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Colin La, Dimitris Gahtidis, Jeremy Rabinowicz, Jurgen Leitner, Yusuke Miyashita","submitted_at":"2023-11-28T10:52:35Z","abstract_excerpt":"In this paper, we propose the use of generative artificial intelligence (AI) to improve zero-shot performance of a pre-trained policy by altering observations during inference. Modern robotic systems, powered by advanced neural networks, have demonstrated remarkable capabilities on pre-trained tasks. However, generalizing and adapting to new objects and environments is challenging, and fine-tuning visuomotor policies is time-consuming. To overcome these issues we propose Robotic Policy Inference via Synthetic Observations (ROSO). ROSO uses stable diffusion to pre-process a robot's observation "},"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":"2311.16680","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2023-11-28T10:52:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cfaf07af994917b4676184e0b0cdb0cb37715bd737bff4d490f9805505c13a14","abstract_canon_sha256":"c427e89e69c7875d455ad303b4c1ab5499b1d24280bf228c80ba171d66d3d8ab"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:17:56.704172Z","signature_b64":"ygSx0qakS5Hi8+eLpX+qFEN4nVA3rUf3QmsCSR17M7dcaQDuzBjLihK5XM3+contdcrxQxTuwt/rLgYg4KWSBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cd02795b036a55b6043d12734228d217348e3cf75c4f769330f656e05f4e8a81","last_reissued_at":"2026-07-05T07:17:56.703744Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:17:56.703744Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ROSO: Improving Robotic Policy Inference via Synthetic Observations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Colin La, Dimitris Gahtidis, Jeremy Rabinowicz, Jurgen Leitner, Yusuke Miyashita","submitted_at":"2023-11-28T10:52:35Z","abstract_excerpt":"In this paper, we propose the use of generative artificial intelligence (AI) to improve zero-shot performance of a pre-trained policy by altering observations during inference. Modern robotic systems, powered by advanced neural networks, have demonstrated remarkable capabilities on pre-trained tasks. However, generalizing and adapting to new objects and environments is challenging, and fine-tuning visuomotor policies is time-consuming. To overcome these issues we propose Robotic Policy Inference via Synthetic Observations (ROSO). ROSO uses stable diffusion to pre-process a robot's observation "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.16680","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/2311.16680/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":"2311.16680","created_at":"2026-07-05T07:17:56.703804+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.16680v2","created_at":"2026-07-05T07:17:56.703804+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.16680","created_at":"2026-07-05T07:17:56.703804+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZUBHSWYDNJK3","created_at":"2026-07-05T07:17:56.703804+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZUBHSWYDNJK3MBB5","created_at":"2026-07-05T07:17:56.703804+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZUBHSWYD","created_at":"2026-07-05T07:17:56.703804+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.16565","citing_title":"Reimagination with Test-time Observation Interventions: Distractor-Robust World Model Predictions for Visual Model Predictive Control","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZUBHSWYDNJK3MBB5CJZUEKGSC4","json":"https://pith.science/pith/ZUBHSWYDNJK3MBB5CJZUEKGSC4.json","graph_json":"https://pith.science/api/pith-number/ZUBHSWYDNJK3MBB5CJZUEKGSC4/graph.json","events_json":"https://pith.science/api/pith-number/ZUBHSWYDNJK3MBB5CJZUEKGSC4/events.json","paper":"https://pith.science/paper/ZUBHSWYD"},"agent_actions":{"view_html":"https://pith.science/pith/ZUBHSWYDNJK3MBB5CJZUEKGSC4","download_json":"https://pith.science/pith/ZUBHSWYDNJK3MBB5CJZUEKGSC4.json","view_paper":"https://pith.science/paper/ZUBHSWYD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.16680&json=true","fetch_graph":"https://pith.science/api/pith-number/ZUBHSWYDNJK3MBB5CJZUEKGSC4/graph.json","fetch_events":"https://pith.science/api/pith-number/ZUBHSWYDNJK3MBB5CJZUEKGSC4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZUBHSWYDNJK3MBB5CJZUEKGSC4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZUBHSWYDNJK3MBB5CJZUEKGSC4/action/storage_attestation","attest_author":"https://pith.science/pith/ZUBHSWYDNJK3MBB5CJZUEKGSC4/action/author_attestation","sign_citation":"https://pith.science/pith/ZUBHSWYDNJK3MBB5CJZUEKGSC4/action/citation_signature","submit_replication":"https://pith.science/pith/ZUBHSWYDNJK3MBB5CJZUEKGSC4/action/replication_record"}},"created_at":"2026-07-05T07:17:56.703804+00:00","updated_at":"2026-07-05T07:17:56.703804+00:00"}