{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:B4O3VRGPZCAOWW5G7F7HY2FLV3","short_pith_number":"pith:B4O3VRGP","schema_version":"1.0","canonical_sha256":"0f1dbac4cfc880eb5ba6f97e7c68abaeeacba40212717298e7bac67da3454c4c","source":{"kind":"arxiv","id":"2605.20223","version":1},"attestation_state":"computed","paper":{"title":"Why Latent Actions Fail, and How to Prevent It","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jung Min Lee, Jungwoo Lee, Li Zhao, Taehyun Cho","submitted_at":"2026-05-13T09:54:35Z","abstract_excerpt":"Latent action models (LAMs) aim to learn action-like representations from unlabeled videos by compressing frame-to-frame changes. The frames of in-the-wild videos, however, contain not only the agent's own state but exogenous state such as background clutter. Since the exogenous state introduces changes unrelated to actions, it hinders reliable latent action learning. This paper investigates this problem analytically by extending a linear LAM framework to explicitly model exogenous state. Our analysis reveals two insights: (1) minimizing the standard reconstruction objective produces latent ac"},"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":"2605.20223","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T09:54:35Z","cross_cats_sorted":[],"title_canon_sha256":"7a825cbd6dd8aa58ebb1a37663be4a3f4ecbdb74420ab95ae9f11584a4969fc5","abstract_canon_sha256":"ed9eafe2bbe230bc25ab775c41e67560ca6781e0a96bc580bc545431b1195f2b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T00:04:21.697565Z","signature_b64":"SwgS58917Nl+EEoNdCzhOpSzVMTH18RszrEDTGTr4mMQKSwvEXBxYz0mDaZAdNiv9Dbq24WPi8p18nduLJjUAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f1dbac4cfc880eb5ba6f97e7c68abaeeacba40212717298e7bac67da3454c4c","last_reissued_at":"2026-05-21T00:04:21.696944Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T00:04:21.696944Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Why Latent Actions Fail, and How to Prevent It","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jung Min Lee, Jungwoo Lee, Li Zhao, Taehyun Cho","submitted_at":"2026-05-13T09:54:35Z","abstract_excerpt":"Latent action models (LAMs) aim to learn action-like representations from unlabeled videos by compressing frame-to-frame changes. The frames of in-the-wild videos, however, contain not only the agent's own state but exogenous state such as background clutter. Since the exogenous state introduces changes unrelated to actions, it hinders reliable latent action learning. This paper investigates this problem analytically by extending a linear LAM framework to explicitly model exogenous state. Our analysis reveals two insights: (1) minimizing the standard reconstruction objective produces latent ac"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.20223","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/2605.20223/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":"2605.20223","created_at":"2026-05-21T00:04:21.697045+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.20223v1","created_at":"2026-05-21T00:04:21.697045+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.20223","created_at":"2026-05-21T00:04:21.697045+00:00"},{"alias_kind":"pith_short_12","alias_value":"B4O3VRGPZCAO","created_at":"2026-05-21T00:04:21.697045+00:00"},{"alias_kind":"pith_short_16","alias_value":"B4O3VRGPZCAOWW5G","created_at":"2026-05-21T00:04:21.697045+00:00"},{"alias_kind":"pith_short_8","alias_value":"B4O3VRGP","created_at":"2026-05-21T00:04:21.697045+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/B4O3VRGPZCAOWW5G7F7HY2FLV3","json":"https://pith.science/pith/B4O3VRGPZCAOWW5G7F7HY2FLV3.json","graph_json":"https://pith.science/api/pith-number/B4O3VRGPZCAOWW5G7F7HY2FLV3/graph.json","events_json":"https://pith.science/api/pith-number/B4O3VRGPZCAOWW5G7F7HY2FLV3/events.json","paper":"https://pith.science/paper/B4O3VRGP"},"agent_actions":{"view_html":"https://pith.science/pith/B4O3VRGPZCAOWW5G7F7HY2FLV3","download_json":"https://pith.science/pith/B4O3VRGPZCAOWW5G7F7HY2FLV3.json","view_paper":"https://pith.science/paper/B4O3VRGP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.20223&json=true","fetch_graph":"https://pith.science/api/pith-number/B4O3VRGPZCAOWW5G7F7HY2FLV3/graph.json","fetch_events":"https://pith.science/api/pith-number/B4O3VRGPZCAOWW5G7F7HY2FLV3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B4O3VRGPZCAOWW5G7F7HY2FLV3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B4O3VRGPZCAOWW5G7F7HY2FLV3/action/storage_attestation","attest_author":"https://pith.science/pith/B4O3VRGPZCAOWW5G7F7HY2FLV3/action/author_attestation","sign_citation":"https://pith.science/pith/B4O3VRGPZCAOWW5G7F7HY2FLV3/action/citation_signature","submit_replication":"https://pith.science/pith/B4O3VRGPZCAOWW5G7F7HY2FLV3/action/replication_record"}},"created_at":"2026-05-21T00:04:21.697045+00:00","updated_at":"2026-05-21T00:04:21.697045+00:00"}