{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:LS7YXZ3QDT7PVK6D3RCIZR3RTE","short_pith_number":"pith:LS7YXZ3Q","canonical_record":{"source":{"id":"2605.14171","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T22:50:31Z","cross_cats_sorted":["cs.NI"],"title_canon_sha256":"dcea29c6222c44c62557a9ec1140f384eefaf2b9f031f9f87e21491f7846e015","abstract_canon_sha256":"1ad3eecff1db839edbbfdd1d2afe5503c0f1a07d9a7be7086f9ea5456f090a67"},"schema_version":"1.0"},"canonical_sha256":"5cbf8be7701cfefaabc3dc448cc771991479a7c8d90840297122deafcda6a367","source":{"kind":"arxiv","id":"2605.14171","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14171","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14171v1","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14171","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"pith_short_12","alias_value":"LS7YXZ3QDT7P","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"LS7YXZ3QDT7PVK6D","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"LS7YXZ3Q","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:LS7YXZ3QDT7PVK6D3RCIZR3RTE","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14171","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T22:50:31Z","cross_cats_sorted":["cs.NI"],"title_canon_sha256":"dcea29c6222c44c62557a9ec1140f384eefaf2b9f031f9f87e21491f7846e015","abstract_canon_sha256":"1ad3eecff1db839edbbfdd1d2afe5503c0f1a07d9a7be7086f9ea5456f090a67"},"schema_version":"1.0"},"canonical_sha256":"5cbf8be7701cfefaabc3dc448cc771991479a7c8d90840297122deafcda6a367","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:11.354054Z","signature_b64":"hBPRzF+O+7Y31p1bDxbz22/UQfVWpNK4Jj88Vozel52nfbCkEXpX3oEt+sCO/PNKteDH0UheF8BSj5s8yjFWBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5cbf8be7701cfefaabc3dc448cc771991479a7c8d90840297122deafcda6a367","last_reissued_at":"2026-05-17T23:39:11.353327Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:11.353327Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14171","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":"2026-05-17T23:39:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oRYlmq/CAnpqSt8thz9GZnAehkBK5wXz6tdMcwIstq8KfCW+n5U7majhR9OCQuKYt8UZFUl6t8QkJK390Kn0Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-22T15:53:51.794645Z"},"content_sha256":"dfeaa75f545546fa4a2668f3965f4b7258775d6eefd33836f25f6adc669d009a","schema_version":"1.0","event_id":"sha256:dfeaa75f545546fa4a2668f3965f4b7258775d6eefd33836f25f6adc669d009a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:LS7YXZ3QDT7PVK6D3RCIZR3RTE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CSI-JEPA: Towards Foundation Representations for Ubiquitous Sensing with Minimal Supervision","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"CSI-JEPA learns reusable temporal-spectral representations from unlabeled Wi-Fi channel state information through masked prediction of latent features.","cross_cats":["cs.NI"],"primary_cat":"cs.LG","authors_text":"Xuanhao Luo, Yuchen Liu, Zhizhen Li","submitted_at":"2026-05-13T22:50:31Z","abstract_excerpt":"Channel state information (CSI) provides a widely available sensing modality for human and environment perception, but existing CSI sensing models usually rely on task-specific supervised training and require substantial labeled data for each task, device, user, or environment. This limits their scalability in practical deployments where unlabeled CSI is abundant but labeled data is costly to collect. In this paper, we present CSI-JEPA, a self-supervised predictive representation learning framework for label-efficient, multi-task Wi-Fi sensing. CSI-JEPA learns reusable temporal-spectral repres"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"CSI-JEPA improves downstream sensing performance over competitive baselines, achieving up to 10.64 percentage points mean accuracy gain over state-of-the-art supervised Transformer and matched-budget label savings of up to 98.0%.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That representations learned via masked prediction on unlabeled CSI transfer effectively to diverse downstream tasks and deployment settings with only lightweight adapters and no substantial domain shift.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CSI-JEPA learns temporal-spectral representations from unlabeled CSI via masked prediction and achieves up to 10.64 percentage points accuracy gain and 98% label savings on seven real-world Wi-Fi sensing tasks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"CSI-JEPA learns reusable temporal-spectral representations from unlabeled Wi-Fi channel state information through masked prediction of latent features.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"4c9cf433c0c6a28e5096877c39d11d34be5af0545c7bf96eef3e8c29cea1d8fb"},"source":{"id":"2605.14171","kind":"arxiv","version":1},"verdict":{"id":"88ddd031-a22f-485c-afbf-35698d398dcd","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T04:50:16.600166Z","strongest_claim":"CSI-JEPA improves downstream sensing performance over competitive baselines, achieving up to 10.64 percentage points mean accuracy gain over state-of-the-art supervised Transformer and matched-budget label savings of up to 98.0%.","one_line_summary":"CSI-JEPA learns temporal-spectral representations from unlabeled CSI via masked prediction and achieves up to 10.64 percentage points accuracy gain and 98% label savings on seven real-world Wi-Fi sensing tasks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That representations learned via masked prediction on unlabeled CSI transfer effectively to diverse downstream tasks and deployment settings with only lightweight adapters and no substantial domain shift.","pith_extraction_headline":"CSI-JEPA learns reusable temporal-spectral representations from unlabeled Wi-Fi channel state information through masked prediction of latent features."},"references":{"count":39,"sample":[{"doi":"","year":2021,"title":"Deep AI enabled ubiquitous wireless sensing: A survey,","work_id":"cb417187-134c-4dff-aa78-f51a3eaebcaa","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Cross-domain WiFi sensing with channel state information: A survey,","work_id":"bdc5f5da-2886-4aaf-99be-55aea64071bf","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"WiFi sensing with channel state information: A survey,","work_id":"1d19cd56-f0ff-4e47-9f4d-da24a3a75d7d","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Wi-Fi can do more: Toward ubiquitous wireless sensing,","work_id":"050a46e0-f949-4b2e-a8b6-64a071e07f33","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"Wireless sensing for human activity: A survey,","work_id":"741dbe10-a010-4b54-91a1-f2fb717fdeb4","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":39,"snapshot_sha256":"398263f17de67851501bb5457188db1e39ad2146a37f9492c011f6982a7aa6de","internal_anchors":1},"formal_canon":{"evidence_count":2,"snapshot_sha256":"537ac396b7005a9293081172972fef7992017edb0e9768bead347f8649e1a31d"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"88ddd031-a22f-485c-afbf-35698d398dcd"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"g62apTGW9nMaGNX7uwOSUsYaoECmbC25w/Tk9k4p6QUTdXmpm3P5G0zlykBeYfqCfEJElfRdNkOyJ+vtDUIqDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-22T15:53:51.795594Z"},"content_sha256":"b3934ebd92e2cded5ecb6039664fe0b25520f08af0a9261f2776a28320a52697","schema_version":"1.0","event_id":"sha256:b3934ebd92e2cded5ecb6039664fe0b25520f08af0a9261f2776a28320a52697"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LS7YXZ3QDT7PVK6D3RCIZR3RTE/bundle.json","state_url":"https://pith.science/pith/LS7YXZ3QDT7PVK6D3RCIZR3RTE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LS7YXZ3QDT7PVK6D3RCIZR3RTE/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":"2026-05-22T15:53:51Z","links":{"resolver":"https://pith.science/pith/LS7YXZ3QDT7PVK6D3RCIZR3RTE","bundle":"https://pith.science/pith/LS7YXZ3QDT7PVK6D3RCIZR3RTE/bundle.json","state":"https://pith.science/pith/LS7YXZ3QDT7PVK6D3RCIZR3RTE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LS7YXZ3QDT7PVK6D3RCIZR3RTE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:LS7YXZ3QDT7PVK6D3RCIZR3RTE","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":"1ad3eecff1db839edbbfdd1d2afe5503c0f1a07d9a7be7086f9ea5456f090a67","cross_cats_sorted":["cs.NI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T22:50:31Z","title_canon_sha256":"dcea29c6222c44c62557a9ec1140f384eefaf2b9f031f9f87e21491f7846e015"},"schema_version":"1.0","source":{"id":"2605.14171","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14171","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14171v1","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14171","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"pith_short_12","alias_value":"LS7YXZ3QDT7P","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"LS7YXZ3QDT7PVK6D","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"LS7YXZ3Q","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:b3934ebd92e2cded5ecb6039664fe0b25520f08af0a9261f2776a28320a52697","target":"graph","created_at":"2026-05-17T23:39:11Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"CSI-JEPA improves downstream sensing performance over competitive baselines, achieving up to 10.64 percentage points mean accuracy gain over state-of-the-art supervised Transformer and matched-budget label savings of up to 98.0%."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That representations learned via masked prediction on unlabeled CSI transfer effectively to diverse downstream tasks and deployment settings with only lightweight adapters and no substantial domain shift."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"CSI-JEPA learns temporal-spectral representations from unlabeled CSI via masked prediction and achieves up to 10.64 percentage points accuracy gain and 98% label savings on seven real-world Wi-Fi sensing tasks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"CSI-JEPA learns reusable temporal-spectral representations from unlabeled Wi-Fi channel state information through masked prediction of latent features."}],"snapshot_sha256":"4c9cf433c0c6a28e5096877c39d11d34be5af0545c7bf96eef3e8c29cea1d8fb"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"537ac396b7005a9293081172972fef7992017edb0e9768bead347f8649e1a31d"},"paper":{"abstract_excerpt":"Channel state information (CSI) provides a widely available sensing modality for human and environment perception, but existing CSI sensing models usually rely on task-specific supervised training and require substantial labeled data for each task, device, user, or environment. This limits their scalability in practical deployments where unlabeled CSI is abundant but labeled data is costly to collect. In this paper, we present CSI-JEPA, a self-supervised predictive representation learning framework for label-efficient, multi-task Wi-Fi sensing. CSI-JEPA learns reusable temporal-spectral repres","authors_text":"Xuanhao Luo, Yuchen Liu, Zhizhen Li","cross_cats":["cs.NI"],"headline":"CSI-JEPA learns reusable temporal-spectral representations from unlabeled Wi-Fi channel state information through masked prediction of latent features.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T22:50:31Z","title":"CSI-JEPA: Towards Foundation Representations for Ubiquitous Sensing with Minimal Supervision"},"references":{"count":39,"internal_anchors":1,"resolved_work":39,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Deep AI enabled ubiquitous wireless sensing: A survey,","work_id":"cb417187-134c-4dff-aa78-f51a3eaebcaa","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Cross-domain WiFi sensing with channel state information: A survey,","work_id":"bdc5f5da-2886-4aaf-99be-55aea64071bf","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"WiFi sensing with channel state information: A survey,","work_id":"1d19cd56-f0ff-4e47-9f4d-da24a3a75d7d","year":2019},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Wi-Fi can do more: Toward ubiquitous wireless sensing,","work_id":"050a46e0-f949-4b2e-a8b6-64a071e07f33","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Wireless sensing for human activity: A survey,","work_id":"741dbe10-a010-4b54-91a1-f2fb717fdeb4","year":2019}],"snapshot_sha256":"398263f17de67851501bb5457188db1e39ad2146a37f9492c011f6982a7aa6de"},"source":{"id":"2605.14171","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T04:50:16.600166Z","id":"88ddd031-a22f-485c-afbf-35698d398dcd","model_set":{"reader":"grok-4.3"},"one_line_summary":"CSI-JEPA learns temporal-spectral representations from unlabeled CSI via masked prediction and achieves up to 10.64 percentage points accuracy gain and 98% label savings on seven real-world Wi-Fi sensing tasks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"CSI-JEPA learns reusable temporal-spectral representations from unlabeled Wi-Fi channel state information through masked prediction of latent features.","strongest_claim":"CSI-JEPA improves downstream sensing performance over competitive baselines, achieving up to 10.64 percentage points mean accuracy gain over state-of-the-art supervised Transformer and matched-budget label savings of up to 98.0%.","weakest_assumption":"That representations learned via masked prediction on unlabeled CSI transfer effectively to diverse downstream tasks and deployment settings with only lightweight adapters and no substantial domain shift."}},"verdict_id":"88ddd031-a22f-485c-afbf-35698d398dcd"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:dfeaa75f545546fa4a2668f3965f4b7258775d6eefd33836f25f6adc669d009a","target":"record","created_at":"2026-05-17T23:39:11Z","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":"1ad3eecff1db839edbbfdd1d2afe5503c0f1a07d9a7be7086f9ea5456f090a67","cross_cats_sorted":["cs.NI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T22:50:31Z","title_canon_sha256":"dcea29c6222c44c62557a9ec1140f384eefaf2b9f031f9f87e21491f7846e015"},"schema_version":"1.0","source":{"id":"2605.14171","kind":"arxiv","version":1}},"canonical_sha256":"5cbf8be7701cfefaabc3dc448cc771991479a7c8d90840297122deafcda6a367","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5cbf8be7701cfefaabc3dc448cc771991479a7c8d90840297122deafcda6a367","first_computed_at":"2026-05-17T23:39:11.353327Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:11.353327Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hBPRzF+O+7Y31p1bDxbz22/UQfVWpNK4Jj88Vozel52nfbCkEXpX3oEt+sCO/PNKteDH0UheF8BSj5s8yjFWBA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:11.354054Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14171","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dfeaa75f545546fa4a2668f3965f4b7258775d6eefd33836f25f6adc669d009a","sha256:b3934ebd92e2cded5ecb6039664fe0b25520f08af0a9261f2776a28320a52697"],"state_sha256":"cbb733d53b5a550978f8b7cc72b682447921f42cfd7cdd414ce6475dea5a6809"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"71oi9ULQfJg7x4QMzGvMRpjoWY4ZwyItIfnpmsyvBvsBNJ8xsh/DNrQxqX25gA5x7hNocsC6xUgnEJuQgH16BA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-22T15:53:51.800106Z","bundle_sha256":"49f0cd0db5b79d0d3f5622d838d4956c9c6ad99a2355b87f5707ba0d6325c171"}}