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.
Figures & tables
Figure 1: Across sessions, electrode drift changes the recorded units, and a decoder fit on Day 0 degrades on Day N without target calibration. APST adapts with frozen weights by conditioning on association profiles estimated from a few labeled calibration trials.
Figure 2: APST overview. (a) Target-session calibration: from M labeled trials, the association estimator produces each unit’s profile pu ; the profile conditions the activity signature at site \scriptsize1⃝ through FiLM and concatenation, forming unit identity eu . (b) Streaming decoding: live activity is combined with the cached identity and profile at site \scriptsize2⃝, aggregated across units by set attention with learned slot queries, then decoded by a sliding-window causal transformer.
Method
Target adaptation
M1 ( R2 / lat.)
M2 ( R2 / lat.)
H1 ( R2 / lat.)
Gradient-free target-session adaptation
WF
None
0.34±0.06
0.06
0.06±0.04
0.08
0.16±0.03
0.15
RNN
None
−0.60±0.45
0.03
−0.07±0.23
0.01
0.09±0.18
0.02
SPINT
Unlabeled few-shot
0.66±0.07
0.13
0.26±0.13
0.13
0.29±0.15
0.14
APST (ours)
Labeled few-shot
0.65±0.11
0.15
0.42±0.10
0.09
0.44±0.15
0.15
Gradient-based target adaptation
Table 1: Official FALCON private held-out performance. Entries are mean ± SD R2 and normalized latency (lat.).
Figure 3: Results on DANDI688 and FALCON (shared y-axis: velocity R2 ). (a) DANDI688 held-out sessions at 32 calibration trials. (b) Calibration curves (solid: Sub-C, dashed: Sub-M; blue: APST, pink: RNN-FT). (c) Ablations on DANDI688 held-out and FALCON dev. (d) Conditioning-site ablation on FALCON dev. (c,d) Three-seed means; negative bars are truncated with values annotated.
Method
M1
M2
H1
WF
0.46 ( 0.12 )
0.15 ( 0.09 )
0.20 ( 0.04 )
RNN
0.52 ( 1.12 )
0.20 ( 0.27 )
0.31 ( 0.22 )
SPINT
0.77 ( 0.11 )
0.59 ( 0.33 )
0.47 ( 0.18 )
CycleGAN + WF
0.61 ( 0.18 )
0.32 ( 0.10 )
0.15 ( 0.03 )
NoMAD + WF
0.64 ( 0.15 )
0.35 ( 0.15 )
0.21 ( 0.08 )
NDT2 Multi
0.77 ( 0.18 )
0.63 ( 0.20 )
0.62 ( 0.10 )
Table 2: Official private held-in mean R2 . Parentheses denote held-in minus held-out gap ( HI−HO ). Bold indicates our method.
Sub-C
11-13
11-16
11-17
11-19
11-20
12-01
APST
0.80
0.78
0.82
0.87
0.83
0.58
Act-only
0.56
0.58
0.65
−0.02
0.46
0.06
RNN-FT
0.83
0.81
0.83
0.83
0.85
0.48
Sub-M
06-23
06-25
06-26
APST
0.84
0.79
0.80
Act-only
0.49
0.60
0.66
Table 3: Held-out velocity R2 at 32 calibration trials. Act-only: Activity-only (no target labels); Dates in 2015 (same sessions as Fig. 3 , different seeds).
Achieving stable long-term neural decoding in invasive brain-machine interfaces (BMIs) remains challenging due to variations in recorded neural populations across sessions. Current latent alignment approaches may overlook task-dependent structure during cross-session adaptation. We propose Task-Conditioned Latent Alignment (TCLA), a framework that stabilizes neural decoding by learning a shared latent space. TCLA learns a low-dimensional source representation using neural reconstruction and continuous behavioral supervision. During target-session adaptation, the shared representation is fixed, while target neural activity is mapped into the source latent space by aligning source and target distributions separately for each task condition. We evaluated TCLA on seven nonhuman primate datasets spanning multiple tasks. In long-term cross-session evaluation, TCLA achieved a mean R2 of 0.476±0.014 with a negative R2 failure rate of only 6.8%. Across 1,356 within-subject session pairs, TCLA achieved a mean R2 of 0.371±0.009 with a failure rate of 6.8%. Across 2,134 cross-subject session pairs, TCLA achieved a mean R2 of 0.218±0.004 with a failure rate of 12.9%, substantially better than those of the comparison methods. These results demonstrate that by preserving behaviorally relevant and task-dependent latent structure, TCLA improves the robustness of neural decoding across recording sessions and subjects. The source code is publicly available at https://github.com/FAMD-CASIA/TCLA.
Canyang Zhao, Bolin Peng, J. Patrick Mayo +2
Institute of Automation, Chinese Academy of Sciences, Beijing, China · Departments of Ophthalmology and Bioengineering, University of Pittsburgh, Pittsburgh, USA · School of Science and Engineering, Chinese University of Hong Kong, Shenzhen, China
Intracortical brain-computer interfaces suffer from day-to-day neural signal shifts that degrade pretrained decoders. Existing unsupervised adaptation methods rely on deep recurrent or adversarial architectures that are too computationally expensive for implantable hardware. We propose Membrane Potential Alignment (MPA), a test-time adaptation method for spiking neural networks that realigns a pretrained decoder to shifted recordings by only matching membrane potential distributions via KL divergence. By restricting updates to low-rank (LoRA) weights, MPA adapts fewer than 9% of parameters. On a non-human primate reaching task spanning over one month, MPA achieves performance competitive with the state-of-the-art NoMAD method, while using a simpler architecture and finer temporal resolution (4 ms vs. 20 ms). These results show that efficient SNN-based test-time adaptation is a practical path toward long-term, recalibration-free brain-computer interfaces.
Guangzhi Tang
Department of Advanced Computing Sciences, Maastricht University Maastricht, The Netherlands
Current high-performing intracortical speech neuroprostheses achieve low word error rates but typically rely on external language models during inference, increasing memory, computation, and latency. In this work, we investigate whether meaningful character-level decoding is achievable without such models. We propose an end-to-end Conformer-based neural decoder trained directly on intracortical recordings from a participant with amyotrophic lateral sclerosis (ALS). Without any external language model, the system achieves a character error rate (CER) of 23.80% on held-out validation data. Analysis shows that performance variability is driven by inter-session signal degradation, while dominant errors arise from incorrect word boundary segmentation. These results demonstrate that effective character-level decoding is possible in a fully end-to-end framework, providing a strong neural signal for downstream linguistic processing.
Owais Mujtaba Khanday, Jose A. Gonzalez-Lopez, Marc Ouellet +2
1Dpt. of Signal Theory, Telematics and Communications, University of Granada, Spain · 2Brain, Mind, and Behavior Research Center, University of Granada, Spain · 3Grupo CSUR de Epilepsia Refractaria, Hospital Virgen de Las Nieves (Granada), Spain