{"paper":{"title":"Bridging Brain and Semantics: A Hierarchical Framework for Semantically Enhanced fMRI-to-Video Reconstruction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"CineNeuron reconstructs videos from fMRI signals through bottom-up semantic enrichment followed by top-down memory integration.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Biao Gong, Chenglong Ma, Chenhui Wang, Hangjie Yuan, Hongming Shan, Jianxiong Gao, Shiwei Zhang, Shuai Tan, Yujie Wei","submitted_at":"2026-05-14T08:39:42Z","abstract_excerpt":"Reconstructing dynamic visual experiences as videos from functional magnetic resonance imaging (fMRI) is pivotal for advancing the understanding of neural processes. However, current fMRI-to-video reconstruction methods are hindered by a semantic gap between noisy fMRI signals and the rich content of videos, stemming from a reliance on incomplete semantic embeddings that neither capture video-specific cues (e.g., actions) nor integrate prior knowledge. To this end, we draw inspiration from the dual-pathway processing mechanism in human brain and introduce CineNeuron, a novel hierarchical frame"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Extensive experimental results on two fMRI-to-video benchmarks demonstrate that CineNeuron surpasses state-of-the-art methods across various metrics.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that the proposed bottom-up semantic enrichment and top-down memory integration stages can reliably capture video-specific cues such as actions without post-hoc tuning or benchmark-specific overfitting.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CineNeuron improves fMRI-to-video reconstruction by combining bottom-up semantic enrichment with top-down Mixture-of-Memories integration and outperforms prior methods on benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"CineNeuron reconstructs videos from fMRI signals through bottom-up semantic enrichment followed by top-down memory integration.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"6b0cb2669fc7d3ae159d3421a5ad7e7764c761391ca8a4ab6f1eceb1d4e7ad4e"},"source":{"id":"2605.14569","kind":"arxiv","version":1},"verdict":{"id":"4c07728f-194f-48cb-a06a-26747f83e7f8","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T01:49:20.522330Z","strongest_claim":"Extensive experimental results on two fMRI-to-video benchmarks demonstrate that CineNeuron surpasses state-of-the-art methods across various metrics.","one_line_summary":"CineNeuron improves fMRI-to-video reconstruction by combining bottom-up semantic enrichment with top-down Mixture-of-Memories integration and outperforms prior methods on benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that the proposed bottom-up semantic enrichment and top-down memory integration stages can reliably capture video-specific cues such as actions without post-hoc tuning or benchmark-specific overfitting.","pith_extraction_headline":"CineNeuron reconstructs videos from fMRI signals through bottom-up semantic enrichment followed by top-down memory integration."},"references":{"count":119,"sample":[{"doi":"","year":null,"title":"A massive 7T fMRI dataset to bridge cognitive neuroscience and ar- tificial intelligence.Nature Neuroscience, 25(1):116–126,","work_id":"bae53ded-9735-4408-a0d4-d248fe7c373e","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"arXiv preprint arXiv:2306.16934 (2023)","work_id":"3754e7d4-269d-41a6-b9f3-7c94d2e69e99","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Signal quality as Achilles’ heel of graph the- ory in functional magnetic resonance imaging in multiple sclerosis.Scientific Reports, 11(1):7376, 2021","work_id":"4763b89f-7fca-4067-ba78-6a1a6d405128","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"arXiv preprint arXiv:2310.19812 , year=","work_id":"4d5cb73b-5ad7-40fa-b46f-12c916a02a0e","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2005,"title":"Structure and func- tion of visual area MT.Annu","work_id":"87fb425d-0217-49d8-87bc-6e1ffeee53b5","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":119,"snapshot_sha256":"99d26de8ce3ddf1b76af12bbf98df19cb1f1e4415a9fcff9262aed3fbe1c47f9","internal_anchors":18},"formal_canon":{"evidence_count":2,"snapshot_sha256":"2708e52662d2edcac9732a0619a3227a419729846f245236a006097592e756b2"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}