{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:LTHAQG6YSIG35US7QVRFYDFWFV","short_pith_number":"pith:LTHAQG6Y","canonical_record":{"source":{"id":"2603.25629","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-03-26T16:41:59Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"14834bc4dfcc77c58201a4e649a7ed23adf42dad802eb471aa9f6ceea056107d","abstract_canon_sha256":"3ad58b80fc03cc320da50707fac017de926edb34df828a464240c8e03ead76e2"},"schema_version":"1.0"},"canonical_sha256":"5cce081bd8920dbed25f85625c0cb62d7e7b786f4680048cc557e04d3248ed2d","source":{"kind":"arxiv","id":"2603.25629","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.25629","created_at":"2026-07-28T01:22:32Z"},{"alias_kind":"arxiv_version","alias_value":"2603.25629v2","created_at":"2026-07-28T01:22:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.25629","created_at":"2026-07-28T01:22:32Z"},{"alias_kind":"pith_short_12","alias_value":"LTHAQG6YSIG3","created_at":"2026-07-28T01:22:32Z"},{"alias_kind":"pith_short_16","alias_value":"LTHAQG6YSIG35US7","created_at":"2026-07-28T01:22:32Z"},{"alias_kind":"pith_short_8","alias_value":"LTHAQG6Y","created_at":"2026-07-28T01:22:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:LTHAQG6YSIG35US7QVRFYDFWFV","target":"record","payload":{"canonical_record":{"source":{"id":"2603.25629","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-03-26T16:41:59Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"14834bc4dfcc77c58201a4e649a7ed23adf42dad802eb471aa9f6ceea056107d","abstract_canon_sha256":"3ad58b80fc03cc320da50707fac017de926edb34df828a464240c8e03ead76e2"},"schema_version":"1.0"},"canonical_sha256":"5cce081bd8920dbed25f85625c0cb62d7e7b786f4680048cc557e04d3248ed2d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T01:22:32.159996Z","signature_b64":"qoY8ceGLVxff8od3ASZ6h9Q6PkMBYL7Zx6Q51ZaYJA62O953037ZLd2SdM0jfiWvtMiQmNGmYmkg7q6TgbaQAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5cce081bd8920dbed25f85625c0cb62d7e7b786f4680048cc557e04d3248ed2d","last_reissued_at":"2026-07-28T01:22:32.159065Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T01:22:32.159065Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2603.25629","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-28T01:22:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VPV0fGZUnnhgmFhg0pfBuVm0nG7RpySrYPN5Z+Lkiu27BaSzCVNSu/n/DReV2+dCqWeZY4F2LBN/yb67iyLSBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T12:53:39.433918Z"},"content_sha256":"50c91afef8b1960a81cb37309ded5ddd909eb4f60d40392d44f5e72ebbedb732","schema_version":"1.0","event_id":"sha256:50c91afef8b1960a81cb37309ded5ddd909eb4f60d40392d44f5e72ebbedb732"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:LTHAQG6YSIG35US7QVRFYDFWFV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LanteRn: Latent Visual Structured Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Andr\\'e G. Viveiros, Andr\\'e Martins, Matthias Lindemann, Nuno Gon\\c{c}alves","submitted_at":"2026-03-26T16:41:59Z","abstract_excerpt":"While language reasoning models excel in many tasks, visual reasoning remains challenging for current large multimodal models (LMMs). As a result, most LMMs default to verbalizing perceptual content into text, a strong limitation for tasks requiring fine-grained spatial and visual understanding. While recent approaches take steps toward thinking with images by invoking tools or generating intermediate images, they either rely on external modules, or incur unnecessary computation by reasoning directly in pixel space. In this paper, we introduce LanteRn, a framework that enables LMMs to interlea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.25629","kind":"arxiv","version":2},"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/2603.25629/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-28T01:22:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/dAfASYYUcec5XFpKP2yE9/WlJ4FZHzfK6kTt3a22XBx4U5KrXUH7zqt4bcdYSS1QBv83v48ue46OPLe34n9CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T12:53:39.435275Z"},"content_sha256":"0fd6f3bb13e6da8a1824c180ab96b9835238359eb44880ac861dec4f928b1ebc","schema_version":"1.0","event_id":"sha256:0fd6f3bb13e6da8a1824c180ab96b9835238359eb44880ac861dec4f928b1ebc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LTHAQG6YSIG35US7QVRFYDFWFV/bundle.json","state_url":"https://pith.science/pith/LTHAQG6YSIG35US7QVRFYDFWFV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LTHAQG6YSIG35US7QVRFYDFWFV/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":"2026-08-05T12:53:39Z","links":{"resolver":"https://pith.science/pith/LTHAQG6YSIG35US7QVRFYDFWFV","bundle":"https://pith.science/pith/LTHAQG6YSIG35US7QVRFYDFWFV/bundle.json","state":"https://pith.science/pith/LTHAQG6YSIG35US7QVRFYDFWFV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LTHAQG6YSIG35US7QVRFYDFWFV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:LTHAQG6YSIG35US7QVRFYDFWFV","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":"3ad58b80fc03cc320da50707fac017de926edb34df828a464240c8e03ead76e2","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-03-26T16:41:59Z","title_canon_sha256":"14834bc4dfcc77c58201a4e649a7ed23adf42dad802eb471aa9f6ceea056107d"},"schema_version":"1.0","source":{"id":"2603.25629","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.25629","created_at":"2026-07-28T01:22:32Z"},{"alias_kind":"arxiv_version","alias_value":"2603.25629v2","created_at":"2026-07-28T01:22:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.25629","created_at":"2026-07-28T01:22:32Z"},{"alias_kind":"pith_short_12","alias_value":"LTHAQG6YSIG3","created_at":"2026-07-28T01:22:32Z"},{"alias_kind":"pith_short_16","alias_value":"LTHAQG6YSIG35US7","created_at":"2026-07-28T01:22:32Z"},{"alias_kind":"pith_short_8","alias_value":"LTHAQG6Y","created_at":"2026-07-28T01:22:32Z"}],"graph_snapshots":[{"event_id":"sha256:0fd6f3bb13e6da8a1824c180ab96b9835238359eb44880ac861dec4f928b1ebc","target":"graph","created_at":"2026-07-28T01:22:32Z","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"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2603.25629/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While language reasoning models excel in many tasks, visual reasoning remains challenging for current large multimodal models (LMMs). As a result, most LMMs default to verbalizing perceptual content into text, a strong limitation for tasks requiring fine-grained spatial and visual understanding. While recent approaches take steps toward thinking with images by invoking tools or generating intermediate images, they either rely on external modules, or incur unnecessary computation by reasoning directly in pixel space. In this paper, we introduce LanteRn, a framework that enables LMMs to interlea","authors_text":"Andr\\'e G. Viveiros, Andr\\'e Martins, Matthias Lindemann, Nuno Gon\\c{c}alves","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-03-26T16:41:59Z","title":"LanteRn: Latent Visual Structured Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.25629","kind":"arxiv","version":2},"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":"sha256:50c91afef8b1960a81cb37309ded5ddd909eb4f60d40392d44f5e72ebbedb732","target":"record","created_at":"2026-07-28T01:22:32Z","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":"3ad58b80fc03cc320da50707fac017de926edb34df828a464240c8e03ead76e2","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-03-26T16:41:59Z","title_canon_sha256":"14834bc4dfcc77c58201a4e649a7ed23adf42dad802eb471aa9f6ceea056107d"},"schema_version":"1.0","source":{"id":"2603.25629","kind":"arxiv","version":2}},"canonical_sha256":"5cce081bd8920dbed25f85625c0cb62d7e7b786f4680048cc557e04d3248ed2d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5cce081bd8920dbed25f85625c0cb62d7e7b786f4680048cc557e04d3248ed2d","first_computed_at":"2026-07-28T01:22:32.159065Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-28T01:22:32.159065Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qoY8ceGLVxff8od3ASZ6h9Q6PkMBYL7Zx6Q51ZaYJA62O953037ZLd2SdM0jfiWvtMiQmNGmYmkg7q6TgbaQAQ==","signature_status":"signed_v1","signed_at":"2026-07-28T01:22:32.159996Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.25629","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:50c91afef8b1960a81cb37309ded5ddd909eb4f60d40392d44f5e72ebbedb732","sha256:0fd6f3bb13e6da8a1824c180ab96b9835238359eb44880ac861dec4f928b1ebc"],"state_sha256":"b28c584150eb8cf944eacfa06623d009d8f7f4f61c136228d6eede5811f45dbf"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"36yuI6piuaKUdf6W13+JAdpWr77BaWZ1X5TCeBVwEzII4cpKg+fo18enzdKLj88chb5a05KqWxHm55uW/JXZDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T12:53:39.452747Z","bundle_sha256":"d6fac6eedc3c2f33a27f503704f301c6d0290d16a0b2bae22cfa132ccc79e583"}}