NeurIPS 2026 论文列表

High-throughput scientific sensing, exemplified by genomics, faces a critical representation bottleneck: discrete biological events are often obscured by continuous, high-entropy stochastic noise. Standard neural codecs fail to distinguish deterministic semantics from stochastic sensor noise, inefficiently allocating their information budget to high-entropy fluctuations rather than to low-entropy structural motifs. We introduce \textsc{WaveSem}, a frequency-adaptive framework that imposes a physically motivated inductive bias on discrete representation learning. Through integrated wavelet decomposition, \textsc{WaveSem} explicitly disentangles each signal into a structured semantic core and a residual texture component. We validate this approach on large-scale nanopore sequencing data, a domain characterized by extreme sequence lengths and low signal-to-noise ratios (SNRs). \textsc{WaveSem} achieves a $14\times$ storage reduction while maintaining competitive downstream basecalling accuracy. Moreover, linear probing of the frozen encoder reaches $\mathrm{AUROC}=0.928$ for 5mC methylation detection, demonstrating that the learned tokens preserve semantic information beyond nucleotide identity. These results establish disentangled vector-quantized representations as a scalable foundation for efficient genomic analysis and long-term archival.

This work proposes WaveGen, a fully end-to-end diffusion-based text-to-waveform framework based on Spectral Forcing, which jointly learns an internal spectral trajectory and the target waveform trajectory within a single model, enabling direct end-to-end text-to-waveform generation. Unlike recent TTS pipelines, WaveGen does not rely on separately trained or externally decoded intermediate representations, such as neural audio codecs, Mel-spectrogram vocoders, or VAE-based latent models. In addition, WaveGen performs internal text-speech alignment within a single model, eliminating the need for external alignment modules such as duration predictors. To preserve a truly end-to-end formulation, our framework further avoids dependence on self-supervised representations, such as BERT-based text embeddings and Wav2Vec 2.0-based semantic representations, as well as semantic distillation methods for training optimization. To achieve this, we carefully design Spectral Forcing, a model architecture with disentangled diffusion heads, and a training objective based on a multi-scale spectral $v$-loss. We further validate the scalability of end-to-end TTS models by scaling the model size from 0.1B to 0.4B parameters. With a single architecture, open-source data, and one-stage training without external modules, WaveGen demonstrates fully end-to-end waveform generation by achieving promising performance. In particular, WaveGen achieves a WER of 1.77, a SIM of 0.63, and competitive MOS performance on the Seed-en benchmark.

Room acoustics modeling requires capturing intricate wave phenomena beyond direct sound propagation, including reflections, refractions, and diffractions. Recent neural acoustic synthesis methods have progressively incorporated richer scene representations and physics-informed priors, yet typically learn acoustic behavior without considering wave propagation dynamics, which inherently capture diffraction and interference phenomena. We propose WavNAF, a framework that bridges wave dynamics and neural acoustic modeling by leveraging Finite-Difference Time-Domain (FDTD) simulations, which solve the wave equation on discrete spatiotemporal grids to capture these phenomena. However, directly applying FDTD at full scale is computationally prohibitive under CFL stability constraints. Inspired by traditional acoustic scale models, WavNAF constructs a geometry-preserving, scaled digital replica from NeRF-derived scene geometry and acoustically probes it with temporally compressed FDTD simulation. The resulting pressure maps are encoded not as direct full-scale Room Impulse Response (RIR) estimates, but as compressed-domain wave signatures that provide the neural acoustic field with a structured inductive bias beyond static geometry. To bridge the mismatch between these scaled simulation features and full-scale acoustic responses, we introduce a Neural Acoustic Scaling Module that adaptively recalibrates pressure-map features at the feature level for full-scale RIR prediction. Experiments show that WavNAF yields consistent gains over prior neural baselines across standard acoustic metrics.

Dynamic Graph Neural Networks (DyGNNs) demonstrate powerful representation capabilities in crucial time-series systems by leveraging graph structure and temporal dynamics. However, existing dynamic graph models often suffer from unpredictable distribution shifts induced by complex time-varying factors such as emergency incidents, leading to severe performance degradation due to their low out-of-distribution (OOD) generalization. Graph invariant learning has been extensively studied to solve this problem through feature disentanglement in the time or spectral domain. Despite recent advances, existing methods suffer from critical limitations in practice: Temporal-based methods struggle to disentangle transient perturbations and stable patterns that are highly overlapping and entangled within the same period. While spectral-based methods attempt spectral-domain disentanglement via the global Fourier transform, these methods inherently lose temporal localization and fail to effectively characterize distribution shifts driven by time-varying factors. To this end, in this paper, we propose **Wav**elet **C**oefficient-Domain **I**nvariant **L**earning **(WavCIL)**, a novel framework for resolving distribution shifts by conducting disentanglement within a wavelet coefficient domain endowed with time-frequency localization capabilities. Specifically, WavCIL consists of two key components: (i) a wavelet transform module, which leverages a set of wavelet bases to map the input temporal signals into wavelet coefficient domain, simultaneously capturing frequency components and localizing their temporal occurrences; and (ii) a coefficient domain invariant learning module, which utilizes an adaptive mask mechanism to disentangle stable causal patterns invariant to time-varying factors from spurious patterns driven by time-varying factors in the coefficient domain, guiding the prediction process to rely more on the causal patterns. Extensive experiments on several benchmark datasets demonstrate that WavCIL achieves state-of-the-art generalization performance when handling distribution shifts. Our code is released at https://anonymous.4open.science/r/WavCIL-4B8C.

Recent advances in latent diffusion models have enabled high-quality image generation, but also raise critical concerns for intellectual property protection in model distribution scenarios, where downstream users have unrestricted access to models, allowing arbitrary modifications. Existing watermarking methods either rely on inference-time control of inputs (e.g., specific prompts or noise initialization) or embed watermark signals in auxiliary components, making them easily removable in such settings. In this paper, we propose INMARK, which treats watermarking as an intrinsic property learned by the model. Instead of relying on input control or auxiliary watermarking components, INMARK enables the core denoising network to internalize and reproduce watermark patterns. Extensive experiments demonstrate that INMARK achieves strong generation fidelity, high watermark detectability, and robustness against attacks, while remaining fully compatible with standard diffusion training pipelines. Our results highlight a new perspective on diffusion model watermarking: the denoising network can learn a reliable and persistent watermarking capability, which is crucial in practical model distribution scenarios.

Watermarking has emerged as a leading technical proposal for attributing generative AI content and is increasingly cited in global governance frameworks. This position paper argues that current implementations risk serving as symbolic compliance rather than delivering effective oversight. We identify a growing gap between regulatory expectations and the technical limitations of existing watermarking schemes. Through analysis of policy proposals and industry practices, we show how incentive structures disincentivize robust, auditable deployments. To realign watermarking with governance goals, we propose a three-layer framework encompassing technical standards, audit infrastructure, and enforcement mechanisms. Without enforceable requirements and independent verification, watermarking will remain inadequate for accountability and ultimately undermine broader efforts in AI safety and regulation.

Watermarking techniques for large language models (LLMs), which encode hidden information in the output so its source can be verified, have gained significant attention in recent days, thanks to their potential capability to detect accidental or deliberate misuse. Similar challenges involving model misuse also exist in the context of game-playing, such as when detecting the unauthorized use of AI tools in gaming platforms (e.g., cheating in online chess). In this paper, we initiate the study of how game-playing strategies can be watermarked. We show how the KGW watermark for LLMs can be adapted to watermark game-playing agents in perfect-information extensive-form games. The watermark can then be detected using a statistical test. We show that the degradation in the quality of the watermarked strategy profile, quantified by the expected utility, can be bounded, but there is a tradeoff between detectability and quality. In our experiments, we bootstrap the watermarking framework to various chess engines and demonstrate that a) the impact of the watermark on the quality of the strategy is negligible and b) the watermark can be detected with just a handful of games.

Video reasoning has a distinctive structure: in a chain of thought over video, most tokens follow from prior text, but a small minority hinge on specific moments of the visual input---and which tokens those are varies from one example to the next. Current post-training procedures like GRPO supply supervision that is coarse in exactly this dimension, applying a single scalar reward uniformly across the trajectory while requiring many rollouts and/or costly external judge models. Our key observation is that a model can audit its own visual reliance: at any position, the shift in its next-token distribution when the video is removed from context measures how much that prediction actually used what was seen. Computing this quantity for both a student and a privileged teacher yields a per-token \emph{reliance gap} that localizes positions at which the teacher's predictions depend on the video and the student's do not. We introduce Visually-Aware On-Policy Self-Distillation (VIOS), which turns this self-auditing signal into dense per-token supervision recovered from a single rollout of a single model---no external reward, judge, or larger teacher required. VIOS combines three components: an exponential moving average (EMA) teacher that co-evolves with the student to avoid the ceiling of a frozen rationalizer; a contrastive term that increases the student's sensitivity to the video input; and a reliance gap gate that concentrates supervision on the positions identified. Across 12 benchmarks spanning general, temporal, spatial, knowledge, and long-video understanding, VIOS consistently outperforms its base model and contemporary video-reasoning systems while training in over $9\times$ fewer steps.

Reconstructing population dynamics is a central problem in the physical and data sciences. Often, the dynamics are modeled as a Wasserstein gradient flow (WGF): a curve of distributions driven by an energy functional. Though there are multiple mathematical characterizations of a WGF, the dominant algorithmic approach relies on the Jordan--Kinderlehrer--Otto (JKO) scheme. JKO-based methods are inflexible to time discretisation and require solving costly optimal transport problems. We take a residual approach, enforcing that the continuity equation conditions are satisfied via a non-negative loss function, whose minimum is the WGF. Combined with a data-fitting divergence, this gives a single global objective. This perspective unifies several existing methods, and leads us to discover a new particle-based method, \emph{stitching}, that is simulation-free and robust to large gaps in observation times of the dynamics. We demonstrate that the stitching method achieves state-of-the-art performance across trajectory inference benchmarks.

There has been a plethora of algorithms proposed that leverage Wasserstein gradient flows or forward-only diffusion processes to perform sampling tasks. These approaches are often characterized by theoretical guarantees of exponentially fast convergence to the target distribution. In this work, we study the mixing behavior of this family of sampling methods. By invoking the Jordan--Kinderlehrer--Otto scheme and Otto calculus, we first establish that Wasserstein gradient flow and forward diffusion-based samplers share the same density evolution. Consequently, their convergence can be characterized by studying the convergence behavior of the associated diffusion process, which has been extensively analyzed in nonequilibrium statistical physics. We perform two independent and well-established analyses for this purpose, namely spectral analysis and mean first-passage time analysis. We show that, although the sampling distributions converge to the target distribution exponentially fast, the presence of multimodality can lead to exponentially long mixing times associated with small spectral gaps. Further analysis elucidates that even when combined with annealing, this family of samplers still requires impractically long time to sample multimodal distributions. We discuss the origin of this behavior as a consequence of the local (i.e., gradient-driven) dynamics underlying this class of samplers, and highlight the need for non-local mechanisms to facilitate the transport of probability mass across modes, enabling efficient exploration in multimodal settings.

Computing the partition function $\mathcal{Z} = \mathrm{tr}(e^{-\beta H})$ of a quantum spin Hamiltonian $H$ on $n$ sites is a fundamental problem in statistical physics and quantum chemistry. However exact evaluation requires $O(8^n)$ time and $O(4^n)$ memory making it intractable for all but the smallest systems. We propose an unbiased stochastic estimator of $\mathcal{Z}$, and more generally of $\mathrm{tr}(f(H))$ for any function $f$ admitting a convergent power series over exponentially large, implicitly defined operators using only local random walks, avoiding both matrix materialization and matrix-vector products. Our approach combines Graph Random Features with the Hutchinson's stochastic trace estimator, and exploits the algebraic structure of quantum spin Hamiltonians to reduce the per-step computational cost to $O(n + d)$, where $d$ is the mean degree of the spin coupling graph, independent of the Hilbert space dimension $N = 2^n$. We validate the estimator empirically against exact matrix exponentiation for small system sizes, and demonstrate its applicability to system sizes where exact methods are infeasible.

This paper introduces Walk3D, a novel 3D Gaussian Splatting (3DGS) framework explicitly designed for 3D referring segmentation, particularly excelling at interpreting complex, long-sentence descriptions. Our primary motivation stems from observing that existing 3DGS-based open-vocabulary methods mainly focus on simple, category-level object segmentation. These methods struggle to precisely segment targets described by highly detailed, multi-sentence queries due to weak compositional language understanding and a lack of structural scene awareness. To tackle these dense and complex linguistic constraints, we propose a spatial routing network that decodes these queries into sequential chain-of-thought instructions upon a scene graph. Specifically, we construct a static supernode graph based on the geometric and semantic properties of the 3DGS space, where a spatial Graph Neural Network (GNN) dynamically pathfinds using the decomposed linguistic chain to accurately segment the referred target. Extensive experiments on complex 3D referring segmentation benchmarks demonstrate the superiority of our proposed method in interpreting lengthy, multi-conditional queries. Our code and trained models will be publicly released.

Large language model (LLM) agents increasingly operate through multi-step tool-use trajectories, where harmful intent may be distributed across actions that appear benign in isolation. This challenges single-step moderation, while repeated LLM-as-judge evaluation over growing prefixes can be costly and may intervene too late. We introduce *WURI* (*Watching Unfolding Risk in Agent Interactions*), a lightweight monitor for early prefix-level detection of harmful agent trajectories. WURI encodes each observed textual step with a frozen text encoder, learns an atom-adapted trajectory representation through metric learning, and scores each prefix with a fixed prototype-margin rule. It requires no access to agent model weights or hidden states, does not modify the agent's internal model, and uses no external LLM-as-judge calls at inference. Across five generalization settings, WURI achieves the best average prefix-area under the detection curve (AUDC), ranks first on three settings, and reaches the strict early-detection operating point within four steps in multiple evaluation settings. Runtime analysis shows a wall-clock speedup of over two orders of magnitude compared with repeated full-prefix guardrail evaluation. These results demonstrate that representation-based prefix scoring is an effective and efficient direction for monitoring unfolding risk in LLM agent interactions.

Goal recognition is the task of inferring an agent's intended goal from its observed behavior. Existing work has largely focused on high-level symbolic recognition, abstracting away the continuous motion through which agents reach their goal. Methods that do reason over real-world trajectories, rely on an explicit dynamics model that is hard to acquire and rarely transfers across settings. Neither approach explicitly learns how agents move toward each candidate goal, limiting accuracy on ambiguous trajectory prefixes where such knowledge is most needed. We present WoRD (Waypoint Recognition with Diffusion), which uses a learned trajectory prior to infer the agent's goal from partial observations. At each step, WoRD runs one warm-started denoising chain per candidate goal, steers each chain toward the observed trajectory through a closed-form log-likelihood gradient, and reweights the resulting samples by Bayes' rule into a calibrated posterior over goals. We evaluate WoRD on simulated robot settings, pedestrian trajectories, and vehicle trajectories using four metrics targeting distinct failure modes. WoRD is the only method that improves all four metrics jointly, including the safety-critical confident-and-wrong rate. WoRD runs under a single inference procedure and preserves calibrated multimodal. beliefs in complex, ambiguous environments where filter- and classifier-based observers collapse to a single mode or grow overconfident.

Transformer-based time series forecasting models often suffer from block-wise attention collapse, where temporal smoothness and cross-variable correlations concentrate attention locally and restrict global modeling capacity. We argue that this collapse is partly a pre-attention representation-conditioning problem: attention logits inherit the spectral concentration and low effective rank of the input representation. We propose Widely Linear Frequency-Domain (WILD) conditioning, a lightweight, model-agnostic module inserted before self-attention. WILD maps inputs to the frequency domain, applies data-dependent spectral preconditioning to redistribute over-dominant frequency energy, and then uses asymmetric real/imaginary projections to realize a widely linear transform beyond fixed time-domain channel mixing. The preconditioning step is essential: it constructs the usable normalized frequency basis on which the widely linear branch can operate. Thus, WILD uses frequency-domain modeling to condition an existing attention layer, not to replace it with a new forecasting architecture. Extensive experiments on eight benchmark datasets show positive gains on 105 of 112 model-dataset-metric configurations across seven backbones. Attention diagnostics show that WILD substantially increases effective rank over the baseline, and a W_r=W_i ablation identifies spectral preconditioning as the primary mechanism, with widely linear mixing providing a secondary, dataset-dependent gain.

The advancement of speech technology has predominantly favored high-resource languages, creating a significant digital divide for speakers of most Sub-Saharan African languages. To address this gap, we introduce WAXAL, a large-scale, openly accessible speech dataset for 24 languages (spanning 19 for ASR and 13 for TTS, with overlaps) representing over 100 million speakers. The collection consists of an Automatic Speech Recognition (ASR) dataset containing approximately 2,340 hours of transcribed, image-prompted natural speech (over 14,100 hours of total collected audio) from a diverse range of speakers, and a Text-to-Speech (TTS) dataset with over 235 hours of high-quality, single-speaker studio recordings of phonetically balanced scripts. Crucially, this effort was executed in partnership with four African academic and community organizations, with the explicit goal of building the local ecosystem for speech technology and seeding data collection capabilities within these institutions. To demonstrate the dataset's utility, we benchmark three architecturally distinct ASR models---Gemma 3n-2B, Whisper-Large-v3, and MMS-1b-all---across the 19 ASR languages. Fine-tuning on WAXAL yields substantial improvements, reducing the macro-average Word Error Rate (WER) by up to 61\% (e.g., from 1.23 to 0.48 for Whisper). Models fine-tuned solely on WAXAL also generalize to the independently collected FLEURS benchmark, confirming the dataset's value for domain-transferable ASR adaptation. Furthermore, we employ an LLM-as-judge framework to evaluate semantic meaning preservation, demonstrating that fine-tuned predictions successfully capture core intents of native speakers. The WAXAL datasets are released at https://huggingface.co/datasets/google/WaxalNLP under the CC-BY-4.0 license to catalyze research and the development of inclusive speech technologies for speakers of African languages.

Magnetic Resonance Imaging (MRI) is widely regarded as the optimal sensor for fetal brain analysis due to its superior soft tissue contrast and anatomical detail. However, its high cost and operational burden make it invasive and difficult to obtain at scale. Ultrasound (US), in contrast, is cheap, safe, and routinely acquired, and as a result it has produced substantially larger datasets and a growing ecosystem of pretrained models. This asymmetry raises a natural question: Can we teach a US-only model to understand fetal MRI from only a limited set of examples? The standard recipe, training a foundation model on subject-to-subject paired MRI-US scans, is not viable since no such paired fetal dataset is publicly available. In this paper we address this gap with Weakly Aligned Spatiotemporal Pairs (WASP), a framework that formulates cross-modal correspondence as an entropic Optimal Transport problem driven by clinical metadata — in particular, Gestational Age (GA) and diagnostic planes — enabling the fitting of a lightweight alignment head that lifts MRI representations into the US latent space, without fine-tuning the backbone. Empirically, WASP yields its largest gains when MRI is unseen by the model during pretraining (on USFM, GA estimation error drops from $21.9$ to $16.3$ days and standard plane classification accuracy climbs from 61.9\% to 74.5\%), while still providing refining improvements for backbones pretrained on both modalities (e.g., BioMedParse GA estimation error from $6.5$ to $5.6$ days and SAM-Med2d plane accuracy from $83.3$\% to $88.1$\%).

Recent advances in brain decoding have achieved impressive visual reconstruction under subject-specific settings, yet they fail to generalize to unseen subjects without retraining---hindering practical zero-shot cross-subject applications. The core challenges stem from inter-subject variability in voxel dimensionality and misaligned spatial response patterns. While existing methods rely on surface-based alignment or complex feature disentanglement, they often operate as black boxes and overlook the potential of volume-based fMRI data. To address this, we propose \textbf{Voxel as Token (VoxTok)}, a conceptual framework that treats each fMRI voxel as an independent token endowed with a learnable functional embedding. This formulation naturally handles variable-length inputs and subsumes existing adapter-based models as special cases where linear projections approximate voxel functionality. Guided by the hypothesis that large-scale spatial distributions of neural activity are consistent across subjects despite fine-grained variability, we introduce a KNN-based estimator to predict functional embeddings for unseen subjects. This approach achieves state-of-the-art zero-shot retrieval performance and comparable reconstruction quality, while enabling plug-and-play adaptation for existing models (e.g., MindEye2). Integrating multi-scale ROIs further boosts performance, with our best model achieving $40.4\%$ zero-shot image retrieval accuracy, rivaling early supervised methods. Crucially, we find that decoding success correlates strongly with ROI size, indicating that global co-activation patterns drive cross-subject generalization.

Sparse attention is becoming increasingly important for serving large language models (LLMs) as generation lengths continue to grow. However, deploying and evaluating new sparse attention algorithms at scale remains highly engineering-intensive, slowing both human researchers and AI agents in exploring the sparse attention design. To address this challenge, we present Vortex, a system that combines a Python-embedded frontend language atop a page-centric tensor abstraction for expressing a broad range of sparse attention algorithms, with an efficient backend tightly integrated into modern LLM serving stacks. Vortex enables rapid prototyping, deployment, and evaluation of sparse attention algorithms, effectively translating their theoretical efficiency gains into real-world throughput improvements. As a result, Vortex substantially accelerates the design and iteration of sparse attention algorithms. Using Vortex, we further demonstrate AI-agent-driven exploration of the sparse attention, automatically generating and refining diverse algorithms that achieve strong accuracy-throughput trade-offs, with the best generated algorithms delivering up to $3.46\times$ higher throughput than full attention while preserving accuracy.