{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AF42JF6ORE4K6DU7ZA6O6BOMXG","short_pith_number":"pith:AF42JF6O","schema_version":"1.0","canonical_sha256":"0179a497ce8938af0e9fc83cef05ccb983a0cb1f067f04e90c79940634d31a81","source":{"kind":"arxiv","id":"2411.04986","version":3},"attestation_state":"computed","paper":{"title":"The Semantic Hub Hypothesis: Language Models Share Semantic Representations Across Languages and Modalities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dani Yogatama, Jiasen Lu, Xinyan Velocity Yu, Yoon Kim, Zhaofeng Wu","submitted_at":"2024-11-07T18:55:09Z","abstract_excerpt":"Modern language models can process inputs across diverse languages and modalities. We hypothesize that models acquire this capability through learning a shared representation space across heterogeneous data types (e.g., different languages and modalities), which places semantically similar inputs near one another, even if they are from different modalities/languages. We term this the semantic hub hypothesis, following the hub-and-spoke model from neuroscience (Patterson et al., 2007) which posits that semantic knowledge in the human brain is organized through a transmodal semantic \"hub\" which "},"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":"2411.04986","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-07T18:55:09Z","cross_cats_sorted":[],"title_canon_sha256":"72ecbd8800ac327512fe296f8b8c93a1c7ddd22cc345e25358e6aba701552eff","abstract_canon_sha256":"da7b24f9081f81b60354a3f1965f1b5372d839a300838df9a3b306b85e3c4399"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:28:26.025304Z","signature_b64":"fIO/8/LnIhHrBJeutkBkQquYtif/BZ9YtevjFYNk4qMPVQKOR/8gcN/u+eu/kHwRWgbO359YZMVtZCpDBMxZCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0179a497ce8938af0e9fc83cef05ccb983a0cb1f067f04e90c79940634d31a81","last_reissued_at":"2026-07-05T10:28:26.024663Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:28:26.024663Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Semantic Hub Hypothesis: Language Models Share Semantic Representations Across Languages and Modalities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dani Yogatama, Jiasen Lu, Xinyan Velocity Yu, Yoon Kim, Zhaofeng Wu","submitted_at":"2024-11-07T18:55:09Z","abstract_excerpt":"Modern language models can process inputs across diverse languages and modalities. We hypothesize that models acquire this capability through learning a shared representation space across heterogeneous data types (e.g., different languages and modalities), which places semantically similar inputs near one another, even if they are from different modalities/languages. We term this the semantic hub hypothesis, following the hub-and-spoke model from neuroscience (Patterson et al., 2007) which posits that semantic knowledge in the human brain is organized through a transmodal semantic \"hub\" which "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.04986","kind":"arxiv","version":3},"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/2411.04986/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":"2411.04986","created_at":"2026-07-05T10:28:26.024732+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.04986v3","created_at":"2026-07-05T10:28:26.024732+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.04986","created_at":"2026-07-05T10:28:26.024732+00:00"},{"alias_kind":"pith_short_12","alias_value":"AF42JF6ORE4K","created_at":"2026-07-05T10:28:26.024732+00:00"},{"alias_kind":"pith_short_16","alias_value":"AF42JF6ORE4K6DU7","created_at":"2026-07-05T10:28:26.024732+00:00"},{"alias_kind":"pith_short_8","alias_value":"AF42JF6O","created_at":"2026-07-05T10:28:26.024732+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22473","citing_title":"Interleaved Speech Language Models Latently Work In Text","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00356","citing_title":"How Far Do Auto-Interpretation Labels Generalize: A Controlled Study Across Languages, Scripts, and Rewordings","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01196","citing_title":"Low-Resource Safety Failures Are Action Failures, Not Representation Failures","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2512.12744","citing_title":"Resting Neurons, Active Insights: Robustifying Activation Sparsity in LLMs via Spontaneity","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2512.12744","citing_title":"Resting Neurons, Active Insights: Robustifying Activation Sparsity in LLMs via Spontaneity","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09352","citing_title":"The Wittgensteinian Representation Hypothesis: Is Language the Attractor of Multimodal Convergence?","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AF42JF6ORE4K6DU7ZA6O6BOMXG","json":"https://pith.science/pith/AF42JF6ORE4K6DU7ZA6O6BOMXG.json","graph_json":"https://pith.science/api/pith-number/AF42JF6ORE4K6DU7ZA6O6BOMXG/graph.json","events_json":"https://pith.science/api/pith-number/AF42JF6ORE4K6DU7ZA6O6BOMXG/events.json","paper":"https://pith.science/paper/AF42JF6O"},"agent_actions":{"view_html":"https://pith.science/pith/AF42JF6ORE4K6DU7ZA6O6BOMXG","download_json":"https://pith.science/pith/AF42JF6ORE4K6DU7ZA6O6BOMXG.json","view_paper":"https://pith.science/paper/AF42JF6O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.04986&json=true","fetch_graph":"https://pith.science/api/pith-number/AF42JF6ORE4K6DU7ZA6O6BOMXG/graph.json","fetch_events":"https://pith.science/api/pith-number/AF42JF6ORE4K6DU7ZA6O6BOMXG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AF42JF6ORE4K6DU7ZA6O6BOMXG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AF42JF6ORE4K6DU7ZA6O6BOMXG/action/storage_attestation","attest_author":"https://pith.science/pith/AF42JF6ORE4K6DU7ZA6O6BOMXG/action/author_attestation","sign_citation":"https://pith.science/pith/AF42JF6ORE4K6DU7ZA6O6BOMXG/action/citation_signature","submit_replication":"https://pith.science/pith/AF42JF6ORE4K6DU7ZA6O6BOMXG/action/replication_record"}},"created_at":"2026-07-05T10:28:26.024732+00:00","updated_at":"2026-07-05T10:28:26.024732+00:00"}