NeurIPS 2026 论文列表

This position paper argues that {\sl fair representations cannot hold what they promise}. Fair representation learning has been a significant trend in research on fairness for machine learning models over the past decade. The main idea of fair representation learning is to provide a data preprocessing mechanism which can facilitate fairness for downstream tasks, and research papers in this area often insinuate a fairness guarantee for all downstream tasks without further qualification. However, provably this cannot be satisfied by any non-degenerate representation. We argue that such over-promising can be harmful in downstream applications and that the responsibility for fairness cannot safely be outsourced to a representation provider in the way current research seems to suggest. As a minimal requirement for a way forward, we suggest that newly proposed fairness representations should be accompanied by compatibility tests that would allow a user to verify whether the representation will actually guarantee fairness for the task at hand.

The Strong Platonic Representation Hypothesis suggests that representational convergence in artificial neural networks can be harnessed constructively: embeddings can be translated across models through a universal latent space without paired data. We ask whether an analogous geometry can be recovered across human brains. Using fMRI data from the Natural Scenes Dataset, we propose a self-supervised encoder that learns subject-specific embeddings from brain data alone by exploiting repeated stimulus presentations. We show that these independently learned spaces can be translated across subjects using unsupervised orthogonal rotations, without paired cross-subject samples or intermediate model representations. Synchronizing pairwise rotations into a single shared latent space further improves cross-subject retrieval, indicating that subject-specific spaces are mutually compatible with a common coordinate system. These results provide evidence for a shared neural geometry in the human visual cortex: subject-specific fMRI representations are approximately isometric across individuals and can be translated through purely geometric transformations.

We propose a physics-informed method for solving high-dimensional optimal control problems that requires only control labels, not value function labels. A neural network parameterizes the value function, and the control policy is derived from its gradient. The network is trained by minimizing a weighted sum of the standard Hamilton–Jacobi–Bellman (HJB) residual and the mismatch between the network’s control output and control labels, which are precomputed by a state-dependent Riccati equation (SDRE) solver. Theoretically, with the help of the viscous HJB equation, we establish error bounds for the value function and its gradient, providing a rigorous performance guarantee for the resulting policy. Numerical experiments show that the learned model accurately approximates the optimal value function, closely matches the cost-to-go of the SDRE reference, and yields a neural controller that is significantly faster while exhibiting excellent stability.

Real-time jet tagging is critical for identifying short-lived particle decays in the high-throughput detectors of the Large Hadron Collider, where real-time trigger systems which are responsible for deciding which collision events to store impose strict latency and accuracy constraints. Transformer architectures achieve the highest jet tagging accuracy when compute is unconstrained, but their quadratic self-attention cost places them orders of magnitude beyond the trigger budget. Existing efficient variants reduce this cost by compressing the attention matrix or restricting it to ordered local windows, at the price of the explicit particle-particle interactions that drive substructure identification. To address this limitation, we introduce the Patch Hierarchical Attention Transformer (PHAT-JeT), which combines two mechanisms: a physics-inspired geometric message-passing module that encodes local detector-plane structure, and a hierarchical patch-based attention scheme that computes exact attention within small particle groups while preserving global context through lightweight patch-token communication. Within this compute budget, PHAT-JeT achieves state-of-the-art accuracy and background rejection among resource-constrained jet tagging models on four benchmarks (hls4ml, JetClass, Top Tagging, and Quark-Gluon). Our code is available at https://anonymous.4open.science/r/PHAT-JeT-540B/README.md

Representations are routinely used in machine learning, psychology, and neuroscience to probe the computations of biological and artificial systems. Yet it remains unclear to what extent computation constrains representation in artificial neural networks. Neural networks admit *parameter symmetries*: transformations of the parameters that preserve function exactly but reshape representational geometry. We show that known parameter symmetries act on representations through just three primitives: *addition*, *duplication*, and *scaling*, giving a closed-form descriptor of representational geometry as task-linked features plus symmetry-induced noise. This decomposition yields analytic bounds on representational similarity between networks related by parameter symmetry, which reveal when functionally equivalent networks become arbitrarily *dis*similar. Finally, we identify *privileged* representational geometries, which weight features by their computational importance and recover a stable link between representation and computation. Overall, our results delineate when representation can support inferences about computation, and when it cannot.

Precision medicine relies on integrating diverse patient data, including genetic variants, clinical phenotypes, and medical imaging, to inform personalized prevention, diagnosis, and treatment. However, learning general patient-level representations from these heterogeneous modalities remains challenging due to their high dimensionality and pervasive missingness. We propose PM1, a multimodal foundation model that learns meaningful patient-level representations across genetic, clinical, and imaging data from 337,129 individuals in the UK Biobank, with a focus on retinal imaging and ophthalmic traits. To address modality dimensionality, PM1 adopts an intermediate fusion architecture with modality-specific encoders and a shared Transformer backbone, where a novel pathway-based genotype encoder enables the incorporation of over 600,000 genetic variants by reducing dimensionality while preserving trait-relevant variation. In addition, a contrastive loss invariant to modality missingness allows PM1 to learn robust patient-level representations despite only 6\% of UK Biobank participants having complete modality coverage. By partitioning modalities into interpretable tokens, PM1 enables interpretability methods to investigate known and potentially novel associations across genetic, clinical, and imaging features, and their interactions. Evaluated across downstream tasks, including phenotype prediction, conditioned image generation, and attribution analyses, PM1 demonstrates the value of multimodal patient representations, improving predictive performance over unimodal baselines and competing methods while enabling interpretable analyses in settings with heterogeneous and incomplete patient data.

Characterizing the features of a Hamiltonian that governs a quantum system serves as a fundamental subroutine of quantum device calibration, signal sensing, and error correction. Recent works proposed protocols have achieved the optimal Heisenberg-limited scaling learning ansatz-free Hamiltonians from their real-time evolutions without fully specifying interaction structures. However, these protocols relies on both deep circuits with interleaving probes and control, and extremely short time resolution, making them difficult to implement on near- and intermediate-term in situ quantum experiments. In this work, we propose a computational efficient, *control-free*, and *ancilla-free* algorithm that uses only *Pauli product state preparation and measurement*, and learns an ansatz-free Hamiltonian $H$ with $||H||\leq\Lambda$ in total evolution time of $\Theta(\tfrac{\Lambda}{\epsilon^2}\log(\tfrac{\Lambda}{\epsilon}))$. The evolution time cost of our algorithm is *optimal* for any control-free protocols as we further prove a lower bound of $\Omega(\tfrac{\Lambda}{\epsilon^2}\log(\tfrac{\Lambda}{\epsilon}))$. Technically, our method introduces a randomized-sampling framework that combines band-limited kernel-based time sampling with a displacement sieve for Hamiltonian structure learning. The characteristic probe time resolution depends only on $\Lambda$ instead of $\varepsilon$, which makes our protocol especially appealing in the high-precision regime for sensing and calibration applications.

Understanding model predictions is essential for physical applications, where outputs often inform safety-critical decisions, such as structural load assessment, weather warnings, and clinical diagnosis. Shapley values satisfy many desirable properties as an attribution method, but their computational cost during inference hinders their practical use. Current amortized explainers, such as FastSHAP, are limited to homogeneous inputs, which is problematic for physical applications where data often comes from irregular grids and geometries. We introduce OperatorSHAP, a grid-agnostic attribution method and training procedure that allows us to train FastSHAP-like explainers for neural operators. We establish a theoretical framework for attributions in function space, connecting to Aumann–Shapley values. We further show that OperatorSHAP's explanations are consistent with state-of-the-art discrete Shapley values across resolutions and transfer across grid sizes without retraining.

Mobile and wearable devices offer an unprecedented opportunity for continuous, passive health monitoring and active health coaching. However, the largest wearable datasets are not publicly available for research, and leading wearable foundation models trained on such datasets are rarely open-weight or come with reproducible training code. To accelerate open science in wearable health, we release **OpenMyHeartCounts (OpenMHC)**, the largest and most comprehensive open-access wearable health dataset to-date, alongside open-source implementations of recent wearable foundation models. \openmhc{}, derived from over a decade of data collected through the My Heart Counts study app, includes >60 million hours of wearable data across 19 sensor channels (e.g., step count, heart rate, sleep, workouts) and up to 169 linked variables, including health, lifestyle, mood, and behavior from 11,894 consenting participants. Furthermore, we introduce a unified, open benchmark that enables standardized comparison of wearable health models across three tracks: health and behavior downstream prediction, multivariate data imputation, and time-series forecasting. We benchmark classical methods alongside recent wearable and multivariate time series foundation models. By open-sourcing data, code, and model weights at this unprecedented scale, we aim to democratize wearable health AI research and enable the community to drive open progress in this domain.

Bayesian model calibration is central to digital twins and computer experiments, as it aligns model outputs with field observations by estimating calibration parameters and correcting systematic model bias. Classical Bayesian calibration introduces latent parameters and a discrepancy function to capture bias, but suffers from parameter–discrepancy confounding and is typically formulated as an offline procedure under a stationary data-generating assumption. These limitations are restrictive in modern digital twin applications, where systems evolve over time and may exhibit both gradual drift and abrupt regime shifts. While data assimilation methods enable sequential updates, they generally do not explicitly model systematic bias and are not robust to abrupt changes. We propose Bayesian Recursive Projected Calibration (BRPC), an online Bayesian calibration framework for streaming data under simulator mismatch and nonstationarity. BRPC extends projected calibration to the online setting by separating a discrepancy-free particle update for calibration parameters from a conditional Gaussian process update for discrepancy, preserving identifiability while enabling bias-aware adaptation under gradual system evolution. To handle abrupt changes, we integrate BRPC with restart mechanisms that detect regime shifts and reset the calibration process, yielding a unified approach to mixed nonstationarity. We establish theoretical guarantees for both components, including tracking performance under gradual evolution and false-alarm and detection behavior for restart mechanisms. Empirical results on synthetic and plant-simulation benchmarks demonstrate that BRPC improves calibration accuracy under gradual changes, while the restart-augmented framework further enhances robustness and predictive performance under abrupt regime shifts compared to sliding-window Bayesian calibration and data assimilation baselines.

Backpropagation is the cornerstone of deep learning, but its reliance on symmetric weight transport and global synchronization makes it computationally expensive and biologically implausible. Feedback alignment offers a promising alternative by approximating error gradients through fixed random feedback, thereby avoiding symmetric weight transport. However, this approach often struggles with poor learning performance and instability, especially in deep networks. Here, we show that a one-time soft alignment between forward and feedback weights at initialization enables deep networks to achieve performance comparable to backpropagation, without requiring weight transport during learning. This simple initialization condition guides stable error minimization in the loss landscape, improving network trainability. Spectral analyses further reveal that initial alignment promotes smoother gradient flow and convergence to flatter minima, resulting in better generalization and robustness. Notably, we also find that allowing moderate deviations from exact weight symmetry can improve adversarial robustness compared to standard backpropagation. These findings demonstrate that a simple initialization strategy can enable robust and effective learning in deep networks without dynamic weight transport, providing a resource-efficient alternative to standard backpropagation.

Verifiable Reward (RLVR) has emerged as a promising approach for enhancing the reasoning capabilities of large language models. However, it still remains unclear whether RLVR can extend reasoning capabilities beyond what is already encoded in the base model. In this paper, we study this question through the lens of *primacy bias* — the tendency of models to over-rely on already-learned representations, disproportionately improving performance on problems they already handle well while underexploring harder ones that require departing from those representations. Through our controlled experiments and large-scale analysis, we show that RLVR training on mixed data tends to further sharpen performance on easier problems while yielding only marginal gains on harder ones. We find that although hard problems lead to modest gains overall, they are actually the ones that help most in expanding the model’s reasoning capabilities and pushing the representations. However, prolonged training on these hard problems may also hurt and lead to forgetting on easy problems. These findings help to better understand some of the prior conflicting observations about RLVR, specifically, why on (OOD) domains where the base model performs poorly, improvements on hard problems can expand reasoning; while on well-trained domains, forgetting can reduce performance and mask genuine gains obtained from hard problems. Data and code for reproducing our experiments are available at: https://anonymous.4open.science/r/neurips26_sub-0DE8.

Sentence-level EEG-to-text decoding models typically pad variable-duration neural signals to a fixed-length window, creating a structural confound where segment duration correlates with sentence identity independently of neural content, a problem we term length bias. Despite being acknowledged in prior work, the prevalence, magnitude, and architectural dependence of this effect have never been systematically quantified. We introduce length-matched noise baselines that isolate this confound by replacing only the true signal region with Gaussian noise while preserving zero-padding, and apply them across three English listening EEG datasets under a strict subject-held-out protocol. Evaluating ten sentence classifiers and three generation models based on the current state-of-the-art, we show that reported decoding scores are substantially inflated by length cues across datasets, with the degree of inflation varying markedly by architecture. Among classifiers, BIOT alone is unaffected by length-matched noise. A systematic decomposition of its components identifies Short-Fourier-Transform (STFT) tokenisation as the bigger source of this invariance. We transfer this finding into two new generation variants that largely eliminate encoder-side length bias, enabling a controlled analysis of where residual bias originates. A classification-to-generation evaluation further reveals that leading generation models operate largely as sentence retrieval systems, with a linear classifier recovering nearly all of their reported BLEU-4 scores. Together, these results establish that length bias is a pervasive and previously unquantified failure mode in EEG-to-text evaluation, and provide both diagnostic tools and architectural directions for future work.

Irregular data assimilation is difficult because sparse, off-grid observations do not constrain the same state corrections from one cycle to the next. As the observation geometry changes, the set of supported correction directions changes as well. Many learning-based DA methods infer updates in a fixed latent or control space, even when the current observation layout supports a different set of correction directions. To address this mismatch, we introduce Observability-Constrained DataAssimilation with Control-Space Inference (ObsConDA), which determines those supported correction directions from the current background and observation geometry and infers only their control coefficients. On real surface-station assimilation, ObsConDA gives the best held-out station accuracy and the best full-field reconstruction among the tested classical and learned baselines. The same advantage carries to matched synthetic observations, explicit geometry shifts, and synthetic dynamical systems, supporting the view that the usable correction directions must change from cycle to cycle. Ablations show that the gain comes from adapting those directions themselves rather than from adding a latent corrector alone.

Individual neurons exhibit diverse molecular, morphological, and electrophysiological properties that shape their distinct roles in neural circuits. While these multimodal features are largely shared across individuals, they remain an underutilized source of information in analyzing neural population dynamics. We posit that these shared biological profiles provide a common reference for relating neural dynamics across subjects and recording sessions. Our framework leverages this structure by routing neurons into latent groups based on their biological profiles and analyzing population dynamics through group-level representations. Because these assignments are anchored in biological properties that are consistent across subjects, the learned organization supports transfer to unseen animals without retraining or neuron-wise correspondence. We evaluate our framework on whole-brain calcium imaging in larval zebrafish and large-scale electrophysiology in the mouse visual cortex. Across both species and recording modalities, the framework delivers state-of-the-art performance in both neural decoding and manifold analysis: it achieves superior decoding of behavior and stimulus labels while better preserving the temporal consistency and local geometric structure of latent trajectories on unseen data. Our experiments demonstrate that not only does providing biological features improve baseline performance, but our proposed framework achieves even stronger results by using these properties to explicitly structure the population rather than treating them as auxiliary inputs. Together, these results suggest that organizing neural populations by their multimodal biological features offers a promising framework for generalizing population dynamics across subjects and recording sessions.

Understanding how neural activity encodes behavioral and sensory variables is a key challenge in systems neuroscience. This requires models that go beyond predicting activity from labels and instead support source-conditioned counterfactual editing: given a recorded source response, we should be able to edit one behavioral or stimulus variable and predict how neural activity would change while preserving other observed factors and source-specific variability not explained by the labels. Here, we introduce **Neural-DISCO**, a disentangled representation learning framework for counterfactual modeling of neural activity. Neural-DISCO partitions the latent space into label-specific components, each associated with an observed behavioral or stimulus variable, together with a residual latent that captures variability not supplied through observed labels. Across simulated and real neural recording datasets, we show that this structure enables selective manipulation of individual labels to generate counterfactual neural responses while preserving the remaining content of the original activity. Further, to gain insight into how behavioral and stimulus variables are encoded at single-neuron resolution, we pair this disentangled framework with feature attribution methods that identify which neurons are most important for each label. We find that disentanglement enhances the recovery of known functional cell roles in both synthetic and real data. Together, these results support Neural-DISCO as a framework for controlled counterfactual modeling of neural activity and provide a practical interface for probing how neural populations encode behavior and sensory information.

Neural populations exhibit heterogeneous tuning: Neurons differ in their selectivity, gain, and in how their preferred inputs are distributed across stimulus or latent-variable space. While this heterogeneity has been studied extensively in the context of encoding precision, its implications for learning and generalization are less well-understood. Here, we develop a probabilistic theory that links a neural population's tuning statistics to its inductive bias: in large populations, the tuning statistics induce a kernel and, equivalently, a prior of a generative model over functions. This prior in turn predicts the population's ability to generalize across a distribution of related tasks. Our theory determines the population tuning statistics that minimize the generalization error over a distribution of functions. We further present a normative account of tuning adaptation inspired by our probabilistic approach, showing that tuning changes that maximize the marginal likelihood of observed data systematically reduce generalization error relative to non-adapting populations. Applied to the adaptation of hippocampal spatial representations in a reward learning task, we found that our theory could explain the over-representation of rewarding locations through experience in an environment with a localized reward. We further demonstrate how different components of the population's generative model differentially impact the speed and magnitude of tuning adaptation. Overall, our work proposes a novel connection between tuning variability and a population's ability to learn and to generalize across tasks and develops a Bayesian theory of tuning adaptation in neural populations during learning.

Generating clinical reports that summarize abnormal patterns, diagnostic findings, and clinical interpretations from long-term EEG recordings remains labor-intensive. We present CELM, the first clinical EEG-to-Language foundation model capable of summarizing long-duration, variable-length EEG recordings and performing end-to-end clinical report generation at multiple scales. CELM integrates pretrained EEG foundation models with language models to enable scalable multimodal learning. We curate a large-scale clinical EEG dataset containing 9,922 reports paired with approximately 11,000 hours of EEG recordings from 9,048 patients to train CELM, and release the benchmark with an automated report-structuring pipeline to facilitate future research. Experimental results show that CELM consistently outperforms existing methods across all evaluation settings. Importantly, we further conduct human evaluation with clinical experts, demonstrating that CELM generates reports that are more clinically coherent, diagnostically reliable, and better aligned with expert interpretation. We release our model and benchmark construction pipeline at https://anonymous.4open.science/r/CELM-3AF4.

Fractional differential equations (FDEs) excel at modelling systems with long-range memory, while stochastic differential equations (SDEs) capture inherent randomness. Existing neural differential equation models address either memory (Neural FDEs) or stochasticity (Neural SDEs), but a unified framework combining both with a learnable scalar memory exponent and an efficient adjoint remains absent. We introduce the Neural Fractional Stochastic Differential Equation (Neural FSDE), which learns continuous-time dynamics with both memory and randomness from data via a Caputo FSDE with neural drift, neural diffusion, and a learnable fractional order. To enable scalable training, we derive a discrete adjoint method that computes gradients with respect to the initial state, the drift and diffusion network parameters, and the fractional order, with memory cost independent of the autograd graph depth. We validate the framework on two real-world domains. On the stochastic-fractional COVID-19 model of Bonyah et al., the NeuralFSDE provides the first end-to-end real-data calibration under a rolling-origin forecasting protocol, recovering a fractional order less than 1. On S&P~500 index option pricing, we recast rough Bergomi, rough Heston, Neural SDE, and NeuralFSDE as Caputo-FSDE special cases under a single calibration pipeline; the NeuralFSDE achieves the lowest implied-volatility error across schedules and instruments, with a matched-architecture comparison against the memoryless Neural SDE, isolating the contribution of fractional memory. Code is available at https://anonymous.4open.science/r/NeuralFSDE.