{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:RXK7LJGHY5CTKWN3NV6PVFPS5L","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":"007272008734eb7fde2955edf9e9acab4d5bbbb50e741a3220d75cb6fb0d4ec7","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"q-bio.NC","submitted_at":"2026-05-13T04:55:03Z","title_canon_sha256":"5788eb664e6f5a7e361904febfa65259919bf8445cd2ec8c2fdae61d6ce3fcc0"},"schema_version":"1.0","source":{"id":"2605.12999","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12999","created_at":"2026-05-18T03:09:00Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12999v1","created_at":"2026-05-18T03:09:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12999","created_at":"2026-05-18T03:09:00Z"},{"alias_kind":"pith_short_12","alias_value":"RXK7LJGHY5CT","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"RXK7LJGHY5CTKWN3","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"RXK7LJGH","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:eeee292b946a0a3e766599c8e0ffe974ff0947638b8ffd636339902b360dfcfd","target":"graph","created_at":"2026-05-18T03:09:00Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"A single Mamba forecaster, trained only on next-step spike counts at Neuropixels scale, can deliver both a forecast of upcoming neural population activity and a readout of the animal's behavioral state in one forward pass. Mamba's predicted rates decode mouse choice at 75.7±0.2% trial vote and stimulus side at 66.1±0.6%, outperforming a matched linear decoder on raw spikes by 4-6 pp."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the performance gain arises specifically from the forecasting objective and learned representations rather than from differences in temporal context handling, model capacity, or post-hoc per-session fitting of the linear head."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Mamba forecaster trained on next-step spikes decodes mouse choice at 75.7% and stimulus at 66.1%, beating linear decoding on raw spikes by 4-6 percentage points."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A single Mamba forecaster trained only on next-step spike counts delivers both neural forecasts and improved behavioral decoding in one forward pass."}],"snapshot_sha256":"11b2d42dd197385cd59de1b58aed4796d19c8dd1be63454844ed8db16af7416b"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Closed-loop brain-computer interfaces often require both a forecast of upcoming neural population activity and a readout of the animal's behavioral state. A single Mamba forecaster, trained only on next-step spike counts at Neuropixels scale, can deliver both in one forward pass. A lightweight per-session linear head reading the model's predicted rates decodes behavior better than the same linear classifier reading the raw spike counts, under matched temporal context. We test on the Steinmetz visual-discrimination benchmark, which spans 39 sessions, roughly 27,000 neurons, and 1,994 held-out t","authors_text":"Ash Robbins, David Haussler, Jason Eshraghian, Jesus Gonzalez-Ferrer, Jinghui Geng, John R. Minnick, Kamran Hussain, Mircea Teodorescu, Mohammed A. Mostajo-Radji","cross_cats":["cs.LG"],"headline":"A single Mamba forecaster trained only on next-step spike counts delivers both neural forecasts and improved behavioral decoding in one forward pass.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"q-bio.NC","submitted_at":"2026-05-13T04:55:03Z","title":"Implicit Behavioral Decoding from Next-Step Spike Forecasts at Population Scale"},"references":{"count":13,"internal_anchors":1,"resolved_work":13,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"M. Azabou, V. Arora, V. Ganesh, X. Mao, S. Nachimuthu, M. Mendelson, B. Richards, M. Perich, G. Lajoie, and E. Dyer. A Unified, Scalable Framework for Neural Population Decoding. In NeurIPS, 2023","work_id":"029437f0-78e0-4183-ac85-cada6455327a","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"L. Duncker and M. Sahani. Temporal alignment and latent Gaussian process factor inference in population spike trains. NeurIPS, 2018","work_id":"5a6962f0-f81f-4fc8-ba69-495ce8d33ca6","year":2018},{"cited_arxiv_id":"2312.00752","doi":"","is_internal_anchor":true,"ref_index":3,"title":"Mamba: Linear-Time Sequence Modeling with Selective State Spaces","work_id":"4ee75248-1199-492c-a52f-6661e0f4adff","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"International Brain Laboratory, K. Banga, J. Benson, J. Bhagat, D. Biderman, D. Birman, N. Bonacchi, S. A. Bruijns, K. Buchanan, R. A. A. Campbell, et al. Reproducibility of in vivo electrophysiologic","work_id":"968fb3ca-da42-440b-8a50-3efbf3c9ba1d","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"S. W. Linderman, M. J. Johnson, A. C. Miller, R. P. Adams, D. M. Blei, and L. Paninski. Bayesian Learning and Inference in Recurrent Switching Linear Dynamical Systems. In AISTATS, volume 54, pages 91","work_id":"5241627b-914f-4b89-98fd-e6983f30285f","year":2017}],"snapshot_sha256":"aed086b18103a52052666d312fa01bfdadc4109788d07ab7bd3e43b815d5c15c"},"source":{"id":"2605.12999","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T02:09:04.474403Z","id":"6493ec95-b3c4-4e99-9ed2-e693a93f6f82","model_set":{"reader":"grok-4.3"},"one_line_summary":"Mamba forecaster trained on next-step spikes decodes mouse choice at 75.7% and stimulus at 66.1%, beating linear decoding on raw spikes by 4-6 percentage points.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A single Mamba forecaster trained only on next-step spike counts delivers both neural forecasts and improved behavioral decoding in one forward pass.","strongest_claim":"A single Mamba forecaster, trained only on next-step spike counts at Neuropixels scale, can deliver both a forecast of upcoming neural population activity and a readout of the animal's behavioral state in one forward pass. Mamba's predicted rates decode mouse choice at 75.7±0.2% trial vote and stimulus side at 66.1±0.6%, outperforming a matched linear decoder on raw spikes by 4-6 pp.","weakest_assumption":"That the performance gain arises specifically from the forecasting objective and learned representations rather than from differences in temporal context handling, model capacity, or post-hoc per-session fitting of the linear head."}},"verdict_id":"6493ec95-b3c4-4e99-9ed2-e693a93f6f82"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:53e8fffab9ed1d9ec22726b3daad06d3e235ad80ebcf713733805e50685ae69e","target":"record","created_at":"2026-05-18T03:09:00Z","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":"007272008734eb7fde2955edf9e9acab4d5bbbb50e741a3220d75cb6fb0d4ec7","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"q-bio.NC","submitted_at":"2026-05-13T04:55:03Z","title_canon_sha256":"5788eb664e6f5a7e361904febfa65259919bf8445cd2ec8c2fdae61d6ce3fcc0"},"schema_version":"1.0","source":{"id":"2605.12999","kind":"arxiv","version":1}},"canonical_sha256":"8dd5f5a4c7c7453559bb6d7cfa95f2eada7f707eb24a921142d5aecc75434d9a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8dd5f5a4c7c7453559bb6d7cfa95f2eada7f707eb24a921142d5aecc75434d9a","first_computed_at":"2026-05-18T03:09:00.458029Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:09:00.458029Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qEjBXWLU+eIESFtDp1Bu7o3prRqu47jwSo+NEY6J7G10yOPiJys2Z8xXmlPC1P39JyFH1BXe2ZeG029TzBz3Bw==","signature_status":"signed_v1","signed_at":"2026-05-18T03:09:00.458521Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.12999","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:53e8fffab9ed1d9ec22726b3daad06d3e235ad80ebcf713733805e50685ae69e","sha256:eeee292b946a0a3e766599c8e0ffe974ff0947638b8ffd636339902b360dfcfd"],"state_sha256":"c3245b7634becb81862ca6fa8c8faede673311d7e9bb894bd7919e8c9fb7c542"}