Association profile conditioning in a set-temporal transformer for cross-session intracortical motor decoding
Abstract
Intracortical motor decoders degrade across sessions because the set of recorded units changes and persisting units can alter how their firing relates to behavior. Most existing methods update network weights on each new session or rely on unlabeled activity, which does not directly reveal such changes. We present APST, an Association Profile-conditioned Set-Temporal transformer that adapts to new sessions with all network weights frozen. From a few labeled calibration trials, APST summarizes how each unit's firing relates to behavior in a four-dimensional association profile computed in closed form. The profiles condition a set-attention encoder that accepts any number and order of units, followed by a causal transformer for streaming decoding. On held-out DANDI688 sessions from two monkeys, APST reaches velocity R2 of $0.78$ and $0.81$, versus $0.40$ and $0.58$ for a variant that uses neural activity alone, and matches or exceeds an RNN fine-tuned on the same trials. On FALCON private held-out evaluation, it attains R2 of $0.65$, $0.42$, and $0.44$ on M1, M2, and H1.
Related
- Spatial Sampling and Temporal Dynamics Shape Motor Decoding from Broadband Intracortical Signals
- Real-time intracortical decoding from human motor cortex reveals a task-independent commitment to act that aligns with conscious intent
- Circuit plasticity coordinates action selection and temporal compression during operant conditioning in Aplysia
- FloREN: Decoding Immune Regulatory Networks through Interpretable Graph Transformer Patient Representations
- Striatal dopamine dynamics across temporal scales in brain-specific Atp2a2 heterozygous knockout mice during reward conditioning
BibTeX
@misc{zhang2026association,
title = {{Association profile conditioning in a set-temporal transformer for cross-session intracortical motor decoding}},
author = {Xinyuan Zhang and Handong Mo and Pengfei Wen and Shuang Liang and Jichang Yang and Yan Zeng and Zhongrui Wang and Han Wang},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.39080},
eprint = {2609.39080},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.39080},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Association profile conditioning in a set-temporal transformer for cross-session intracortical motor decoding
AU - Xinyuan Zhang
AU - Handong Mo
AU - Pengfei Wen
AU - Shuang Liang
AU - Jichang Yang
AU - Yan Zeng
AU - Zhongrui Wang
AU - Han Wang
PY - 2026
DA - 2026/09/30
PB - arXiv
DO - 10.48550/arxiv.2609.39080
UR - https://arxiv.org/abs/2609.39080
AB - Intracortical motor decoders degrade across sessions because the set of recorded units changes and persisting units can alter how their firing relates to behavior. Most existing methods update network weights on each new session or rely on unlabeled activity, which does not directly reveal such changes. We present APST, an Association Profile-conditioned Set-Temporal transformer that adapts to new sessions with all network weights frozen. From a few labeled calibration trials, APST summarizes how each unit's firing relates to behavior in a four-dimensional association profile computed in closed form. The profiles condition a set-attention encoder that accepts any number and order of units, followed by a causal transformer for streaming decoding. On held-out DANDI688 sessions from two monkeys, APST reaches velocity R^2 of $0.78$ and $0.81$, versus $0.40$ and $0.58$ for a variant that uses neural activity alone, and matches or exceeds an RNN fine-tuned on the same trials. On FALCON private held-out evaluation, it attains R^2 of $0.65$, $0.42$, and $0.44$ on M1, M2, and H1.
ER -