Q1: Both reviewer #4 and reviewer #5 think it is essential to compare the proposed method with Pre-LayerNorm. We added additional experiments to investigate the question on how PLD compares with PreLN? GLUE score (80.2) compared with Post-LN (82.1) on downstream tasks. When trained with the large learning rate as PLD, PreLN's Q2: Reviewer #3, #4, #5 ask about a comparison to simpler and alternative schedules. The current schedule is actually simple.

Latent factor models have been widely used to analyze simultaneous recordings of spike trains from large, heterogeneous neural populations. These models assume the signal of interest in the population is a low-dimensional latent intensity that evolves over time, which is observed in high dimension via noisy point-process observations. These techniques have been well used to capture neural correlations across a population and to provide a smooth, denoised, and concise representation of high-dimensional spiking data. One limitation of many current models is that the observation model is assumed to be Poisson, which lacks the flexibility to capture under-and over-dispersion that is common in recorded neural data, thereby introducing bias into estimates of covariance. Here we develop the generalized count linear dynamical system, which relaxes the Poisson assumption by using a more general exponential family for count data. In addition to containing Poisson, Bernoulli, negative binomial, and other common count distributions as special cases, we show that this model can be tractably learned by extending recent advances in variational inference techniques. We apply our model to data from primate motor cortex and demonstrate performance improvements over state-of-the-art methods, both in capturing the variance structure of the data and in held-out prediction.

Neural population activity often exhibits rich variability. This variability can arise from single-neuron stochasticity, neural dynamics on short time-scales, as well as from modulations of neural firing properties on long time-scales, often referred to as neural non-stationarity. To better understand the nature of co-variability in neural circuits and their impact on cortical information processing, we introduce a hierarchical dynamics model that is able to capture both slow inter-trial modulations in firing rates as well as neural population dynamics. We derive a Bayesian Laplace propagation algorithm for joint inference of parameters and population states. On neural population recordings from primary visual cortex, we demonstrate that our model provides a better account of the structure of neural firing than stationary dynamics models.

Humans can extract rich semantic information from minimal visual cues, as demonstrated by point-light displays (PLDs), which consist of sparse sets of dots localized to key joints of the human body. This ability emerges early in development and is largely attributed to human embodied experience. Since PLDs isolate body motion as the sole source of meaning, they represent key stimuli for testing the constraints of action understanding in these systems. Here we introduce ActPLD, the first benchmark to evaluate action processing in MLLMs from human PLDs. Tested models include state-of-the-art proprietary and open-source systems on single-actor and socially interacting PLDs. Our results reveal consistently low performance across models, introducing fundamental gaps in action and spatiotemporal understanding.

Submitted by Assigned_Reviewer_1 Q1 This paper extends Poisson linear dynamical systems (PLDS) to account for the non-stationarity in neural spike trains. Their method (NPLDS) uses a hierarchical framework to find the latent variables for each trial, and also scale those latent variables multiplicatively for each trial. The latent variables are found with a linear dynamical system, and the inter-trial modulators are enforced to be smooth across trials with a Gaussian process. To fit the model, the authors devised the Bayesian Laplacian propagation and used an iterative procedure, which may be of interest to those outside the neuroscience field. The results are shown to be more predictive than the previous PLDS method, which suggests the added complexity helps performance.

Q1: Both reviewer #4 and reviewer #5 think it is essential to compare the proposed method with Pre-LayerNorm. We added additional experiments to investigate the question on how PLD compares with PreLN? GLUE score (80.2) compared with Post-LN (82.1) on downstream tasks. When trained with the large learning rate as PLD, PreLN's Q2: Reviewer #3, #4, #5 ask about a comparison to simpler and alternative schedules. The current schedule is actually simple.