{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:ZPGCEW4NA32S4OBQLIEMWDBEV7","short_pith_number":"pith:ZPGCEW4N","schema_version":"1.0","canonical_sha256":"cbcc225b8d06f52e38305a08cb0c24afcf2e17f99e654165478fc54cb619c22f","source":{"kind":"arxiv","id":"2105.03663","version":1},"attestation_state":"computed","paper":{"title":"On Linear Interpolation in the Latent Space of Deep Generative Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Mike Yan Michelis, Quentin Becker","submitted_at":"2021-05-08T10:27:07Z","abstract_excerpt":"The underlying geometrical structure of the latent space in deep generative models is in most cases not Euclidean, which may lead to biases when comparing interpolation capabilities of two models. Smoothness and plausibility of linear interpolations in latent space are associated with the quality of the underlying generative model. In this paper, we show that not all such interpolations are comparable as they can deviate arbitrarily from the shortest interpolation curve given by the geodesic. This deviation is revealed by computing curve lengths with the pull-back metric of the generative mode"},"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":"2105.03663","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-05-08T10:27:07Z","cross_cats_sorted":[],"title_canon_sha256":"39079e6dfe03f00e0ec4dd58f1e7b068f973e476376e43ddfb9af23dd4661dea","abstract_canon_sha256":"d1344f167eb2370d282113757fb973bb42b25fe76d5f54edb72070d479a98a24"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:38:37.763107Z","signature_b64":"tQ7+WPEZFz0vahl298Atks8nIrdp7mIiA45eGwR2pY6EXJHojZNkjZkd+LIuabBasollkfE9UebK00fnWpvMBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cbcc225b8d06f52e38305a08cb0c24afcf2e17f99e654165478fc54cb619c22f","last_reissued_at":"2026-07-05T02:38:37.762701Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:38:37.762701Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Linear Interpolation in the Latent Space of Deep Generative Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Mike Yan Michelis, Quentin Becker","submitted_at":"2021-05-08T10:27:07Z","abstract_excerpt":"The underlying geometrical structure of the latent space in deep generative models is in most cases not Euclidean, which may lead to biases when comparing interpolation capabilities of two models. Smoothness and plausibility of linear interpolations in latent space are associated with the quality of the underlying generative model. In this paper, we show that not all such interpolations are comparable as they can deviate arbitrarily from the shortest interpolation curve given by the geodesic. This deviation is revealed by computing curve lengths with the pull-back metric of the generative mode"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.03663","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/2105.03663/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":"2105.03663","created_at":"2026-07-05T02:38:37.762761+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.03663v1","created_at":"2026-07-05T02:38:37.762761+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.03663","created_at":"2026-07-05T02:38:37.762761+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZPGCEW4NA32S","created_at":"2026-07-05T02:38:37.762761+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZPGCEW4NA32S4OBQ","created_at":"2026-07-05T02:38:37.762761+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZPGCEW4N","created_at":"2026-07-05T02:38:37.762761+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2504.18506","citing_title":"Action-Minimization Meets Generative Modeling: Efficient Transition Path Sampling with the Onsager-Machlup Functional","ref_index":60,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZPGCEW4NA32S4OBQLIEMWDBEV7","json":"https://pith.science/pith/ZPGCEW4NA32S4OBQLIEMWDBEV7.json","graph_json":"https://pith.science/api/pith-number/ZPGCEW4NA32S4OBQLIEMWDBEV7/graph.json","events_json":"https://pith.science/api/pith-number/ZPGCEW4NA32S4OBQLIEMWDBEV7/events.json","paper":"https://pith.science/paper/ZPGCEW4N"},"agent_actions":{"view_html":"https://pith.science/pith/ZPGCEW4NA32S4OBQLIEMWDBEV7","download_json":"https://pith.science/pith/ZPGCEW4NA32S4OBQLIEMWDBEV7.json","view_paper":"https://pith.science/paper/ZPGCEW4N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.03663&json=true","fetch_graph":"https://pith.science/api/pith-number/ZPGCEW4NA32S4OBQLIEMWDBEV7/graph.json","fetch_events":"https://pith.science/api/pith-number/ZPGCEW4NA32S4OBQLIEMWDBEV7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZPGCEW4NA32S4OBQLIEMWDBEV7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZPGCEW4NA32S4OBQLIEMWDBEV7/action/storage_attestation","attest_author":"https://pith.science/pith/ZPGCEW4NA32S4OBQLIEMWDBEV7/action/author_attestation","sign_citation":"https://pith.science/pith/ZPGCEW4NA32S4OBQLIEMWDBEV7/action/citation_signature","submit_replication":"https://pith.science/pith/ZPGCEW4NA32S4OBQLIEMWDBEV7/action/replication_record"}},"created_at":"2026-07-05T02:38:37.762761+00:00","updated_at":"2026-07-05T02:38:37.762761+00:00"}