{"paper":{"title":"Structure Abstraction and Generalization in a Hippocampal-Entorhinal Inspired World Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A hippocampal-entorhinal inspired model abstracts structures from dynamic scenes to enable generalization through path integration.","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.NE","authors_text":"Muyang Lyu, Si Wu, Tianqiu Zhang, Xiao Liu","submitted_at":"2026-05-15T08:36:40Z","abstract_excerpt":"Humans abstract experiences into structured representations to facilitate pattern inference and knowledge transfer. While the hippocampal-entorhinal (HPC-MEC) circuit is known to represent both spatial and conceptual spaces, the mechanisms for concurrently extracting abstract structures from continuous, high-dimensional dynamics remain poorly understood. We propose a brain-inspired hierarchical model that simultaneously infers latent transitions and constructs a predictive visual world model. Our architecture employs an inverse model for structural extraction alongside an HPC-MEC coupling mode"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"By leveraging velocity-driven path integration, the framework enables robust prediction and structural reuse across diverse contexts, thereby achieving structural generalization.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that the proposed inverse model plus HPC-MEC coupling accurately extracts and dissociates relational structures from episodic scenes in a manner that mirrors biological mechanisms and that primitive transformation dynamics constitute a sufficient benchmark for demonstrating this capacity.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A brain-inspired hierarchical model with inverse structural extraction and HPC-MEC dissociation achieves structural abstraction and generalization in visual world models via velocity-driven path integration.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A hippocampal-entorhinal inspired model abstracts structures from dynamic scenes to enable generalization through path integration.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"814bb1c6a39099166e52628c401c6d44d6321023e624d78ffe743add98ff809f"},"source":{"id":"2605.15733","kind":"arxiv","version":1},"verdict":{"id":"2da8e833-38d8-45fe-a339-f3098b971151","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T19:32:39.407279Z","strongest_claim":"By leveraging velocity-driven path integration, the framework enables robust prediction and structural reuse across diverse contexts, thereby achieving structural generalization.","one_line_summary":"A brain-inspired hierarchical model with inverse structural extraction and HPC-MEC dissociation achieves structural abstraction and generalization in visual world models via velocity-driven path integration.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that the proposed inverse model plus HPC-MEC coupling accurately extracts and dissociates relational structures from episodic scenes in a manner that mirrors biological mechanisms and that primitive transformation dynamics constitute a sufficient benchmark for demonstrating this capacity.","pith_extraction_headline":"A hippocampal-entorhinal inspired model abstracts structures from dynamic scenes to enable generalization through path integration."},"integrity":{"clean":false,"summary":{"advisory":2,"critical":0,"by_detector":{"doi_compliance":{"total":2,"advisory":2,"critical":0,"informational":0}},"informational":0},"endpoint":"/pith/2605.15733/integrity.json","findings":[{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.1523/JNEUROSCI.4353-05.2006.URL) was visible in the surrounding text but could not be confirmed against doi.org as printed.","detector":"doi_compliance","severity":"advisory","ref_index":4,"audited_at":"2026-05-19T19:41:02.250764Z","detected_doi":"10.1523/JNEUROSCI.4353-05.2006.URL","finding_type":"recoverable_identifier","verdict_class":"incontrovertible","detected_arxiv_id":null},{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.1002/hipo.20327.Chandra) was visible in the surrounding text but could not be confirmed against doi.org as printed.","detector":"doi_compliance","severity":"advisory","ref_index":2,"audited_at":"2026-05-19T19:41:02.250764Z","detected_doi":"10.1002/hipo.20327.Chandra","finding_type":"recoverable_identifier","verdict_class":"incontrovertible","detected_arxiv_id":null}],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T20:01:19.197194Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T19:41:02.250764Z","status":"completed","version":"1.0.0","findings_count":2},{"name":"ai_meta_artifact","ran_at":"2026-05-19T19:33:24.535262Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T17:21:55.990030Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"50f117e13b8af118dc8128d396c78e700ac275ee383a7c4bf03d1190b2fd6e19"},"references":{"count":18,"sample":[{"doi":"","year":null,"title":"UniVLA: Learning to Act Anywhere with Task-centric Latent Actions","work_id":"e05d654d-db73-48f6-9318-381b6798bac9","ref_index":1,"cited_arxiv_id":"2505.06111","is_internal_anchor":true},{"doi":"10.1002/hipo","year":null,"title":"doi: 10.1002/hipo","work_id":"6415b822-5fae-4c0f-a9a1-722a8aac2aae","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"doi: 10.1038/ s41586-024-08392-y","work_id":"b127536f-15a2-43f9-9edf-3ceb198ec554","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1523/jneurosci","year":2006,"title":"09-07-02382.1989","work_id":"a690fa1b-1b18-47e7-9d31-a8fdbcbb3403","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1038/s41467-021-22559-5","year":2041,"title":"doi: 10.1038/s41467-021-22559-5","work_id":"15e1824c-ee26-446a-9129-8d734d88eabd","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":18,"snapshot_sha256":"a5dbf1ad06cf38408d4cc216560ac13f1499d395d3174581c9a68af686ed3ec4","internal_anchors":3},"formal_canon":{"evidence_count":2,"snapshot_sha256":"589bec9bf4fa17459da5b697bc1dd07a28654d89051c9a4abd061d72607e6522"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}