{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:SUFTN2PQIYEZLEVEAKFN5WR26Y","short_pith_number":"pith:SUFTN2PQ","canonical_record":{"source":{"id":"2610.08432","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-10-06T14:30:42Z","cross_cats_sorted":[],"title_canon_sha256":"d5c8ea68f734cfc82b66df6e41e7d84a90729554c933892884a98a799c4b13ac","abstract_canon_sha256":"81c63f3422211f172c69802ee5d0fbb401407c1111a594ede16694d46f302a91"},"schema_version":"1.0"},"canonical_sha256":"950b36e9f046099592a4028adeda3af60bde9e621133d90f876d237f29d7ce3c","source":{"kind":"arxiv","id":"2610.08432","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2610.08432","created_at":"2026-10-07T01:29:21Z"},{"alias_kind":"arxiv_version","alias_value":"2610.08432v1","created_at":"2026-10-07T01:29:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2610.08432","created_at":"2026-10-07T01:29:21Z"},{"alias_kind":"pith_short_12","alias_value":"SUFTN2PQIYEZ","created_at":"2026-10-07T01:29:21Z"},{"alias_kind":"pith_short_16","alias_value":"SUFTN2PQIYEZLEVE","created_at":"2026-10-07T01:29:21Z"},{"alias_kind":"pith_short_8","alias_value":"SUFTN2PQ","created_at":"2026-10-07T01:29:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:SUFTN2PQIYEZLEVEAKFN5WR26Y","target":"record","payload":{"canonical_record":{"source":{"id":"2610.08432","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-10-06T14:30:42Z","cross_cats_sorted":[],"title_canon_sha256":"d5c8ea68f734cfc82b66df6e41e7d84a90729554c933892884a98a799c4b13ac","abstract_canon_sha256":"81c63f3422211f172c69802ee5d0fbb401407c1111a594ede16694d46f302a91"},"schema_version":"1.0"},"canonical_sha256":"950b36e9f046099592a4028adeda3af60bde9e621133d90f876d237f29d7ce3c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-10-07T01:29:21.202071Z","signature_b64":"YSCuhm4dSBvoswiTMLDn8x3iJFzju/ob1TuMmmmcMNuBWWJdYlnhzf1oIFNSZJi2vQgscvKe3nO4Vo52lkXbCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"950b36e9f046099592a4028adeda3af60bde9e621133d90f876d237f29d7ce3c","last_reissued_at":"2026-10-07T01:29:21.200491Z","signature_status":"signed_v1","first_computed_at":"2026-10-07T01:29:21.200491Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2610.08432","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-10-07T01:29:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zrV2HGUEH+slBrYfJAm+v/cFLSlUm5PPL9MX+C58rV27O/9MboltuPiQOt7CjN3kk+LgUIgQb28tH1EEs5zFBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-11-17T04:58:48.477859Z"},"content_sha256":"627b7dfddbd544a0dfd273cc179ad1281a5dfc9211e9ba5608da6ac811af4921","schema_version":"1.0","event_id":"sha256:627b7dfddbd544a0dfd273cc179ad1281a5dfc9211e9ba5608da6ac811af4921"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:SUFTN2PQIYEZLEVEAKFN5WR26Y","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"EMHO: EMbodied Agent Harness Optimization via Experience Traces","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Donghyun Kim, Hyun Jung Lee, Jongwon Jeong, Jungtaek Kim, Tae-Eui Kam, Yong Jae Lee","submitted_at":"2026-10-06T14:30:42Z","abstract_excerpt":"Improving embodied agents often focuses on optimizing the underlying model through training, while the surrounding agent harness that controls planning, context, and tool use is typically engineered. We ask whether this harness can instead improve itself directly from experience traces under sparse environmental feedback. We propose EMbodied Agent Harness Optimization (EMHO), a self-evolving framework that keeps the embodied model frozen and iteratively revises its harness by analyzing execution trajectories and prior harness history. EMHO optimizes beyond skills or recovery prompts, modifying"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2610.08432","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/2610.08432/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-10-07T01:29:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FOwaRdpIR36oTRtoDwJRdPljR4cSwdvwa6QGTHPq4DMW7yCsoc5lWhQhTRIS0gLOL4WWJHPYZ9+5k8+bE3hwCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-11-17T04:58:48.478457Z"},"content_sha256":"356438777b05aca19d46da4291755210b5316007763abed310ea7bf0ff5644ba","schema_version":"1.0","event_id":"sha256:356438777b05aca19d46da4291755210b5316007763abed310ea7bf0ff5644ba"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SUFTN2PQIYEZLEVEAKFN5WR26Y/bundle.json","state_url":"https://pith.science/pith/SUFTN2PQIYEZLEVEAKFN5WR26Y/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SUFTN2PQIYEZLEVEAKFN5WR26Y/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-11-17T04:58:48Z","links":{"resolver":"https://pith.science/pith/SUFTN2PQIYEZLEVEAKFN5WR26Y","bundle":"https://pith.science/pith/SUFTN2PQIYEZLEVEAKFN5WR26Y/bundle.json","state":"https://pith.science/pith/SUFTN2PQIYEZLEVEAKFN5WR26Y/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SUFTN2PQIYEZLEVEAKFN5WR26Y/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:SUFTN2PQIYEZLEVEAKFN5WR26Y","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"81c63f3422211f172c69802ee5d0fbb401407c1111a594ede16694d46f302a91","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-10-06T14:30:42Z","title_canon_sha256":"d5c8ea68f734cfc82b66df6e41e7d84a90729554c933892884a98a799c4b13ac"},"schema_version":"1.0","source":{"id":"2610.08432","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2610.08432","created_at":"2026-10-07T01:29:21Z"},{"alias_kind":"arxiv_version","alias_value":"2610.08432v1","created_at":"2026-10-07T01:29:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2610.08432","created_at":"2026-10-07T01:29:21Z"},{"alias_kind":"pith_short_12","alias_value":"SUFTN2PQIYEZ","created_at":"2026-10-07T01:29:21Z"},{"alias_kind":"pith_short_16","alias_value":"SUFTN2PQIYEZLEVE","created_at":"2026-10-07T01:29:21Z"},{"alias_kind":"pith_short_8","alias_value":"SUFTN2PQ","created_at":"2026-10-07T01:29:21Z"}],"graph_snapshots":[{"event_id":"sha256:356438777b05aca19d46da4291755210b5316007763abed310ea7bf0ff5644ba","target":"graph","created_at":"2026-10-07T01:29:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2610.08432/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Improving embodied agents often focuses on optimizing the underlying model through training, while the surrounding agent harness that controls planning, context, and tool use is typically engineered. We ask whether this harness can instead improve itself directly from experience traces under sparse environmental feedback. We propose EMbodied Agent Harness Optimization (EMHO), a self-evolving framework that keeps the embodied model frozen and iteratively revises its harness by analyzing execution trajectories and prior harness history. EMHO optimizes beyond skills or recovery prompts, modifying","authors_text":"Donghyun Kim, Hyun Jung Lee, Jongwon Jeong, Jungtaek Kim, Tae-Eui Kam, Yong Jae Lee","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-10-06T14:30:42Z","title":"EMHO: EMbodied Agent Harness Optimization via Experience Traces"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2610.08432","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:627b7dfddbd544a0dfd273cc179ad1281a5dfc9211e9ba5608da6ac811af4921","target":"record","created_at":"2026-10-07T01:29:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"81c63f3422211f172c69802ee5d0fbb401407c1111a594ede16694d46f302a91","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-10-06T14:30:42Z","title_canon_sha256":"d5c8ea68f734cfc82b66df6e41e7d84a90729554c933892884a98a799c4b13ac"},"schema_version":"1.0","source":{"id":"2610.08432","kind":"arxiv","version":1}},"canonical_sha256":"950b36e9f046099592a4028adeda3af60bde9e621133d90f876d237f29d7ce3c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"950b36e9f046099592a4028adeda3af60bde9e621133d90f876d237f29d7ce3c","first_computed_at":"2026-10-07T01:29:21.200491Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-10-07T01:29:21.200491Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YSCuhm4dSBvoswiTMLDn8x3iJFzju/ob1TuMmmmcMNuBWWJdYlnhzf1oIFNSZJi2vQgscvKe3nO4Vo52lkXbCQ==","signature_status":"signed_v1","signed_at":"2026-10-07T01:29:21.202071Z","signed_message":"canonical_sha256_bytes"},"source_id":"2610.08432","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:627b7dfddbd544a0dfd273cc179ad1281a5dfc9211e9ba5608da6ac811af4921","sha256:356438777b05aca19d46da4291755210b5316007763abed310ea7bf0ff5644ba"],"state_sha256":"95be06108cea104b252b28f4e2a298a47f98a9fc1a6071071e8ebfaa05b9da42"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tkCg9yK7sfC3Fyjy6n9Jp2m0VlhnyDIUsyw66L7Hg8UEdCJ9E2WlGvWTGat0KMA2yNxmHemzgckX1H3k8IBmBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-11-17T04:58:48.482649Z","bundle_sha256":"2583f5d048449f181524a3383fe8a7f58468e4065e2f0244adf2c3d4e3e59e58"}}