{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:B6IDQXZ32EA7SFDTHN6EAQGRQ5","short_pith_number":"pith:B6IDQXZ3","schema_version":"1.0","canonical_sha256":"0f90385f3bd101f914733b7c4040d1876c012c58fd56636f1f59748cb800f604","source":{"kind":"arxiv","id":"2509.19658","version":2},"attestation_state":"computed","paper":{"title":"RoboSSM: Scalable In-context Imitation Learning via State-Space Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Bo Liu, Jiaheng Hu, Peter Stone, Qiang Liu, Roberto Mart\\'in-Mart\\'in, Yifeng Zhu, Youngju Yoo","submitted_at":"2025-09-24T00:26:15Z","abstract_excerpt":"In-context imitation learning (ICIL) enables robots to learn tasks from prompts consisting of just a handful of demonstrations. By eliminating the need for parameter updates at deployment time, this paradigm supports few-shot adaptation to novel tasks. However, recent ICIL methods rely on Transformers, which have computational limitations and tend to underperform when handling longer prompts than those seen during training. In this work, we introduce RoboSSM, a scalable recipe for in-context imitation learning based on state-space models (SSM). Specifically, RoboSSM replaces Transformers with "},"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":"2509.19658","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-09-24T00:26:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"de3256646c4ae30717b0397781382799a901269258813122d961f456807466bf","abstract_canon_sha256":"1c5457da3941e66c2911f6b338dcd3878512c54560dd6fc9045025ec33ba8b0e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:12:48.077095Z","signature_b64":"3bIH8vaJEnjkT3EQbFP9Sja0x73x2lnO5xuR7RKWJgNl44h2sPsQJmG5hZmraQ/pMGC9jy8vuYp+Ae6Wi0hqCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f90385f3bd101f914733b7c4040d1876c012c58fd56636f1f59748cb800f604","last_reissued_at":"2026-06-19T16:12:48.076655Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:12:48.076655Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RoboSSM: Scalable In-context Imitation Learning via State-Space Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Bo Liu, Jiaheng Hu, Peter Stone, Qiang Liu, Roberto Mart\\'in-Mart\\'in, Yifeng Zhu, Youngju Yoo","submitted_at":"2025-09-24T00:26:15Z","abstract_excerpt":"In-context imitation learning (ICIL) enables robots to learn tasks from prompts consisting of just a handful of demonstrations. By eliminating the need for parameter updates at deployment time, this paradigm supports few-shot adaptation to novel tasks. However, recent ICIL methods rely on Transformers, which have computational limitations and tend to underperform when handling longer prompts than those seen during training. In this work, we introduce RoboSSM, a scalable recipe for in-context imitation learning based on state-space models (SSM). Specifically, RoboSSM replaces Transformers with "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.19658","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/2509.19658/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":"2509.19658","created_at":"2026-06-19T16:12:48.076714+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.19658v2","created_at":"2026-06-19T16:12:48.076714+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.19658","created_at":"2026-06-19T16:12:48.076714+00:00"},{"alias_kind":"pith_short_12","alias_value":"B6IDQXZ32EA7","created_at":"2026-06-19T16:12:48.076714+00:00"},{"alias_kind":"pith_short_16","alias_value":"B6IDQXZ32EA7SFDT","created_at":"2026-06-19T16:12:48.076714+00:00"},{"alias_kind":"pith_short_8","alias_value":"B6IDQXZ3","created_at":"2026-06-19T16:12:48.076714+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.22671","citing_title":"From Abstraction to Instantiation: Learning Behavioral Representation for Vision-Language-Action Model","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14598","citing_title":"DSSP: Diffusion State Space Policy with Full-History Encoding","ref_index":60,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22671","citing_title":"From Abstraction to Instantiation: Learning Behavioral Representation for Vision-Language-Action Model","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B6IDQXZ32EA7SFDTHN6EAQGRQ5","json":"https://pith.science/pith/B6IDQXZ32EA7SFDTHN6EAQGRQ5.json","graph_json":"https://pith.science/api/pith-number/B6IDQXZ32EA7SFDTHN6EAQGRQ5/graph.json","events_json":"https://pith.science/api/pith-number/B6IDQXZ32EA7SFDTHN6EAQGRQ5/events.json","paper":"https://pith.science/paper/B6IDQXZ3"},"agent_actions":{"view_html":"https://pith.science/pith/B6IDQXZ32EA7SFDTHN6EAQGRQ5","download_json":"https://pith.science/pith/B6IDQXZ32EA7SFDTHN6EAQGRQ5.json","view_paper":"https://pith.science/paper/B6IDQXZ3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.19658&json=true","fetch_graph":"https://pith.science/api/pith-number/B6IDQXZ32EA7SFDTHN6EAQGRQ5/graph.json","fetch_events":"https://pith.science/api/pith-number/B6IDQXZ32EA7SFDTHN6EAQGRQ5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B6IDQXZ32EA7SFDTHN6EAQGRQ5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B6IDQXZ32EA7SFDTHN6EAQGRQ5/action/storage_attestation","attest_author":"https://pith.science/pith/B6IDQXZ32EA7SFDTHN6EAQGRQ5/action/author_attestation","sign_citation":"https://pith.science/pith/B6IDQXZ32EA7SFDTHN6EAQGRQ5/action/citation_signature","submit_replication":"https://pith.science/pith/B6IDQXZ32EA7SFDTHN6EAQGRQ5/action/replication_record"}},"created_at":"2026-06-19T16:12:48.076714+00:00","updated_at":"2026-06-19T16:12:48.076714+00:00"}