{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LX2ZLAUENEMKJO76JZAJGZTCEE","short_pith_number":"pith:LX2ZLAUE","schema_version":"1.0","canonical_sha256":"5df59582846918a4bbfe4e40936662212208c14917e55c71c788c76b0bd3985e","source":{"kind":"arxiv","id":"2605.00271","version":3},"attestation_state":"computed","paper":{"title":"REALM: An RGB- and Event-Aligned Latent Manifold for Cross-Modal Perception","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Event data can be projected into pretrained RGB latent spaces using low-rank adaptation to enable zero-shot application of frozen image decoders to raw event streams.","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.CV","authors_text":"David B. Lindell, Jonathan Kelly, Vincenzo Polizzi","submitted_at":"2026-04-30T22:14:36Z","abstract_excerpt":"Event cameras provide several unique advantages over standard frame-based sensors, including high temporal resolution, low latency, and robustness to extreme lighting. However, existing learning-based approaches for event processing are typically confined to narrow, task-specific silos and lack the ability to generalize across modalities. We address this gap with REALM, a cross-modal framework that learns an RGB- and Event-Aligned Latent Manifold by projecting event representations into the pretrained latent space of RGB foundation models. Instead of task-specific training, we leverage low-ran"},"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":"2605.00271","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-30T22:14:36Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"5a62fa8feda0a0bd9674b6cbd3b19ced6f16d86b8afd15d99e7a1e3a41152ea0","abstract_canon_sha256":"0eb613baf49d2f00ea973a5b1a8fadf322172b6ecd3e808e9a27cc0f643b939d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-02T01:17:31.890193Z","signature_b64":"9svFdvZzfZjSzEvtT7wizm25Nnly9oF3OT4krufPSx2m7/4MOwj8S3Tj/axencoAp0y7PgMwP8DJEUKADpnQAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5df59582846918a4bbfe4e40936662212208c14917e55c71c788c76b0bd3985e","last_reissued_at":"2026-07-02T01:17:31.889775Z","signature_status":"signed_v1","first_computed_at":"2026-07-02T01:17:31.889775Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"REALM: An RGB- and Event-Aligned Latent Manifold for Cross-Modal Perception","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Event data can be projected into pretrained RGB latent spaces using low-rank adaptation to enable zero-shot application of frozen image decoders to raw event streams.","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.CV","authors_text":"David B. Lindell, Jonathan Kelly, Vincenzo Polizzi","submitted_at":"2026-04-30T22:14:36Z","abstract_excerpt":"Event cameras provide several unique advantages over standard frame-based sensors, including high temporal resolution, low latency, and robustness to extreme lighting. However, existing learning-based approaches for event processing are typically confined to narrow, task-specific silos and lack the ability to generalize across modalities. We address this gap with REALM, a cross-modal framework that learns an RGB- and Event-Aligned Latent Manifold by projecting event representations into the pretrained latent space of RGB foundation models. Instead of task-specific training, we leverage low-ran"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"REALM enables the direct, zero-shot application of complex, frozen image-trained decoders, such as MASt3R, to raw event data. We demonstrate state-of-the-art performance in wide-baseline feature matching, significantly outperforming specialized architectures.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That low-rank adaptation of frozen RGB backbones is sufficient to bridge the modality gap and unlock geometric and semantic priors for asynchronous event streams without task-specific training or substantial performance loss.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"REALM aligns event streams with RGB ViT latent spaces via LoRA, enabling zero-shot transfer of image-trained linear heads and complex decoders like MASt3R to raw event data with reported SOTA wide-baseline matching performance.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Event data can be projected into pretrained RGB latent spaces using low-rank adaptation to enable zero-shot application of frozen image decoders to raw event streams.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"0f7c09120682ceb401fcaf098ab4afc60e92e7e2d5ca9cfd62a09bd48ecec67e"},"source":{"id":"2605.00271","kind":"arxiv","version":3},"verdict":{"id":"1cf8bf16-cda0-420f-8f8e-dabfc260067b","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-09T19:43:34.168370Z","strongest_claim":"REALM enables the direct, zero-shot application of complex, frozen image-trained decoders, such as MASt3R, to raw event data. We demonstrate state-of-the-art performance in wide-baseline feature matching, significantly outperforming specialized architectures.","one_line_summary":"REALM aligns event streams with RGB ViT latent spaces via LoRA, enabling zero-shot transfer of image-trained linear heads and complex decoders like MASt3R to raw event data with reported SOTA wide-baseline matching performance.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That low-rank adaptation of frozen RGB backbones is sufficient to bridge the modality gap and unlock geometric and semantic priors for asynchronous event streams without task-specific training or substantial performance loss.","pith_extraction_headline":"Event data can be projected into pretrained RGB latent spaces using low-rank adaptation to enable zero-shot application of frozen image decoders to raw event streams."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.00271/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T20:35:33.042917Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T18:22:28.742328Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"eaa1d9ddc968446da28f0e62239da7fcb3a4de86f6f94078a5d322b8217e5c96"},"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":"2605.00271","created_at":"2026-07-02T01:17:31.889839+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.00271v3","created_at":"2026-07-02T01:17:31.889839+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.00271","created_at":"2026-07-02T01:17:31.889839+00:00"},{"alias_kind":"pith_short_12","alias_value":"LX2ZLAUENEMK","created_at":"2026-07-02T01:17:31.889839+00:00"},{"alias_kind":"pith_short_16","alias_value":"LX2ZLAUENEMKJO76","created_at":"2026-07-02T01:17:31.889839+00:00"},{"alias_kind":"pith_short_8","alias_value":"LX2ZLAUE","created_at":"2026-07-02T01:17:31.889839+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/LX2ZLAUENEMKJO76JZAJGZTCEE","json":"https://pith.science/pith/LX2ZLAUENEMKJO76JZAJGZTCEE.json","graph_json":"https://pith.science/api/pith-number/LX2ZLAUENEMKJO76JZAJGZTCEE/graph.json","events_json":"https://pith.science/api/pith-number/LX2ZLAUENEMKJO76JZAJGZTCEE/events.json","paper":"https://pith.science/paper/LX2ZLAUE"},"agent_actions":{"view_html":"https://pith.science/pith/LX2ZLAUENEMKJO76JZAJGZTCEE","download_json":"https://pith.science/pith/LX2ZLAUENEMKJO76JZAJGZTCEE.json","view_paper":"https://pith.science/paper/LX2ZLAUE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.00271&json=true","fetch_graph":"https://pith.science/api/pith-number/LX2ZLAUENEMKJO76JZAJGZTCEE/graph.json","fetch_events":"https://pith.science/api/pith-number/LX2ZLAUENEMKJO76JZAJGZTCEE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LX2ZLAUENEMKJO76JZAJGZTCEE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LX2ZLAUENEMKJO76JZAJGZTCEE/action/storage_attestation","attest_author":"https://pith.science/pith/LX2ZLAUENEMKJO76JZAJGZTCEE/action/author_attestation","sign_citation":"https://pith.science/pith/LX2ZLAUENEMKJO76JZAJGZTCEE/action/citation_signature","submit_replication":"https://pith.science/pith/LX2ZLAUENEMKJO76JZAJGZTCEE/action/replication_record"}},"created_at":"2026-07-02T01:17:31.889839+00:00","updated_at":"2026-07-02T01:17:31.889839+00:00"}