{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IPZCKIAJW7TNFSVSY35MPAZANX","short_pith_number":"pith:IPZCKIAJ","schema_version":"1.0","canonical_sha256":"43f2252009b7e6d2cab2c6fac783206dfca30db9921b0422e68995368a84479a","source":{"kind":"arxiv","id":"2503.05153","version":2},"attestation_state":"computed","paper":{"title":"Generative Trajectory Stitching through Diffusion Composition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Danfei Xu, Utkarsh A. Mishra, Yilun Du, Yunhao Luo","submitted_at":"2025-03-07T05:22:52Z","abstract_excerpt":"Effective trajectory stitching for long-horizon planning is a significant challenge in robotic decision-making. While diffusion models have shown promise in planning, they are limited to solving tasks similar to those seen in their training data. We propose CompDiffuser, a novel generative approach that can solve new tasks by learning to compositionally stitch together shorter trajectory chunks from previously seen tasks. Our key insight is modeling the trajectory distribution by subdividing it into overlapping chunks and learning their conditional relationships through a single bidirectional "},"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":"2503.05153","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-03-07T05:22:52Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"6f185c7613dcc41541615f55b227a7a0560606b9b8510853b48c9f1dd6ef6873","abstract_canon_sha256":"4ada5c1c4e18d2cdfdd996b3e51997ea001ac2587db5b9f28c89e3bbdbbcc333"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:58:24.320667Z","signature_b64":"ROhRTs9uXxg+As16bAsyeWq32YF3YaYrG3BABIzPnbWMjt72AXNQBTTUdrnL8WrMj9dmvVI9xVmg/bwbCQgaAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"43f2252009b7e6d2cab2c6fac783206dfca30db9921b0422e68995368a84479a","last_reissued_at":"2026-07-05T10:58:24.319870Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:58:24.319870Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generative Trajectory Stitching through Diffusion Composition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Danfei Xu, Utkarsh A. Mishra, Yilun Du, Yunhao Luo","submitted_at":"2025-03-07T05:22:52Z","abstract_excerpt":"Effective trajectory stitching for long-horizon planning is a significant challenge in robotic decision-making. While diffusion models have shown promise in planning, they are limited to solving tasks similar to those seen in their training data. We propose CompDiffuser, a novel generative approach that can solve new tasks by learning to compositionally stitch together shorter trajectory chunks from previously seen tasks. Our key insight is modeling the trajectory distribution by subdividing it into overlapping chunks and learning their conditional relationships through a single bidirectional "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.05153","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/2503.05153/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":"2503.05153","created_at":"2026-07-05T10:58:24.319967+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.05153v2","created_at":"2026-07-05T10:58:24.319967+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.05153","created_at":"2026-07-05T10:58:24.319967+00:00"},{"alias_kind":"pith_short_12","alias_value":"IPZCKIAJW7TN","created_at":"2026-07-05T10:58:24.319967+00:00"},{"alias_kind":"pith_short_16","alias_value":"IPZCKIAJW7TNFSVS","created_at":"2026-07-05T10:58:24.319967+00:00"},{"alias_kind":"pith_short_8","alias_value":"IPZCKIAJ","created_at":"2026-07-05T10:58:24.319967+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07855","citing_title":"NFTR: From Provable Mode-Averaging to Geodesic Subgoal Selection in Offline Goal-Conditioned RL","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2606.21646","citing_title":"Energy-based Compositional Diffusion Planning","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28813","citing_title":"Human2Any: Human-to-Robot Transfer via Constraint-Aware Compositional Planning","ref_index":64,"is_internal_anchor":false},{"citing_arxiv_id":"2506.10137","citing_title":"Self-Predictive Representations for Combinatorial Generalization in Behavioral Cloning","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2509.21983","citing_title":"Hybrid Diffusion for Simultaneous Symbolic and Continuous Planning","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13054","citing_title":"Bridging Domain Gaps with Target-Aligned Generation for Offline Reinforcement Learning","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09999","citing_title":"Muninn: Your Trajectory Diffusion Model But Faster","ref_index":35,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IPZCKIAJW7TNFSVSY35MPAZANX","json":"https://pith.science/pith/IPZCKIAJW7TNFSVSY35MPAZANX.json","graph_json":"https://pith.science/api/pith-number/IPZCKIAJW7TNFSVSY35MPAZANX/graph.json","events_json":"https://pith.science/api/pith-number/IPZCKIAJW7TNFSVSY35MPAZANX/events.json","paper":"https://pith.science/paper/IPZCKIAJ"},"agent_actions":{"view_html":"https://pith.science/pith/IPZCKIAJW7TNFSVSY35MPAZANX","download_json":"https://pith.science/pith/IPZCKIAJW7TNFSVSY35MPAZANX.json","view_paper":"https://pith.science/paper/IPZCKIAJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.05153&json=true","fetch_graph":"https://pith.science/api/pith-number/IPZCKIAJW7TNFSVSY35MPAZANX/graph.json","fetch_events":"https://pith.science/api/pith-number/IPZCKIAJW7TNFSVSY35MPAZANX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IPZCKIAJW7TNFSVSY35MPAZANX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IPZCKIAJW7TNFSVSY35MPAZANX/action/storage_attestation","attest_author":"https://pith.science/pith/IPZCKIAJW7TNFSVSY35MPAZANX/action/author_attestation","sign_citation":"https://pith.science/pith/IPZCKIAJW7TNFSVSY35MPAZANX/action/citation_signature","submit_replication":"https://pith.science/pith/IPZCKIAJW7TNFSVSY35MPAZANX/action/replication_record"}},"created_at":"2026-07-05T10:58:24.319967+00:00","updated_at":"2026-07-05T10:58:24.319967+00:00"}