论文检索

Deep Learning · Generative Models and Autoencoders

Leander Girrbach, Zeynep Akata

Sparse autoencoders (SAEs) are used to analyze embeddings, but their role and practical value are debated. We propose a new perspective on SAEs by demonstrating that they can be naturally understood as topic models. We extend Latent Dirichlet Allocation to embedding spaces and derive the SAE objective as a maximum a posteriori estimator under this model. This view implies SAE features are thematic components rather than steerable directions. Based on this, we introduce SAE-TM, a topic modeling framework that: (1) trains an SAE to learn reusable topic atoms, (2) interprets them as word distributions on downstream data, and (3) merges them into any number of topics without retraining. SAE-TM yields more coherent topics than strong baselines on text and image datasets while maintaining diversity. Finally, we analyze thematic structure in image datasets and trace topic changes over time in Japanese woodblock prints. Our work positions SAEs as effective tools for large-scale thematic analysis across modalities. Code and data will be released upon publication.

Deep Learning · Generative Models and Autoencoders

Daniil Shlenskii, Aleksandr Korotin

Electrostatic generative models such as PFGM++ have recently emerged as a powerful framework, achieving state-of-the-art performance in image synthesis. PFGM++ operates in an extended data space with auxiliary dimensionality $D$, recovering the diffusion model framework as $D\to\infty$, while yielding superior empirical results for finite $D$. Like diffusion models, PFGM++ relies on expensive ODE simulations to generate samples, making it computationally costly. To address this, we propose Inverse Poisson Flow Matching (IPFM), a novel distillation framework that accelerates electrostatic generative models across all values of $D$. Our IPFM reformulates distillation as an inverse problem: learning a generator whose induced electrostatic field matches that of the teacher. We derive a tractable training objective for this problem and show that, as $D \to \infty$, our IPFM closely recovers Score Identity Distillation (SiD), a recent method for distilling diffusion models. Empirically, our IPFM produces distilled generators that achieve near-teacher or even superior sample quality using only a few function evaluations. Moreover, we observe that distillation converges faster for finite $D$ than in the $D \to \infty$ (diffusion) limit, which is consistent with prior findings that finite-$D$ PFGM++ models exhibit more favorable optimization and sampling properties.

Deep Learning · Generative Models and Autoencoders

Qirui Jiao, Daoyuan Chen, Yilun Huang, Xika Lin, Ying Shen, Yaliang Li

While recent Text-to-Image (T2I) models show impressive capabilities in synthesizing images from brief descriptions, they struggle with the long, detailed prompts required for professional applications. We present DetailMaster, a comprehensive benchmark for evaluating T2I capabilities on long prompts with complex compositional requirements, accompanied by an automated data construction pipeline and an evaluation workflow. Comprising expert-validated prompts averaging 284.89 tokens, our benchmark introduces four critical evaluation dimensions: Character Attributes, Structured Character Locations, Multi-Dimensional Scene Attributes, and Spatial/Interactive Relationships. Evaluations on various general-purpose and long-prompt-optimized models reveal critical performance limitations, showing that weak encoders struggle to preserve syntactic dependencies within prompts and diffusion models suffer from attribute leakage under detail-intensive conditions. Through a controlled ablation study under varying constraints, we further show that high-fidelity generation requires a synergistic combination of expanded prompt limits and long-prompt training. We open-source our dataset and code to foster progress in long-prompt-driven T2I generation.

Deep Learning · Generative Models and Autoencoders

Huadai Liu, Wen Wang, Kaicheng Luo, Qian Chen, Xiangang Li, Wei Xue

Continuous Variational Autoencoders (VAEs) serve as the fundamental continuous tokenizer for modern neural audio generation systems, enabling high-fidelity reconstruction while providing a compact, smooth latent space for downstream generative priors. However, continuous VAEs face a fundamental conflict when balancing \textit{compression rate}, \textit{reconstruction fidelity}, and \textit{latent space topology}—a challenge we formalize as the \textbf{Rate-Distortion-Regularity Trilemma}. This trilemma stems from a critical \textit{topological mismatch}: the prevailing isotropic Gaussian prior in standard VAEs imposes a \textit{flat} latent geometry that fails to accommodate audio's \textit{hierarchical} nature, where low-frequency components are structured and compressible while high-frequency components are stochastic and incompressible, leading to \textit{disordered information packing} where crucial semantic features are randomly interleaved with high-entropy noise. To resolve this challenge, we propose \textbf{Structured Topology-Aware Regularization (STAR)}, a general training strategy that reshapes latent space geometry by imposing a growth-based constraint field, routing structural and textural information into channel subspaces with matching capacities. STAR is applicable to any VAE architecture and effectively resolves the trilemma, as demonstrated in CNN-based VAEs. To fully exploit STAR's potential, we present \textbf{STAR-VAE}, combining STAR with a hybrid CNN-Mamba architecture that synergizes local feature extraction with linear-complexity global context modeling, achieving state-of-the-art performance. We further propose \textbf{STAR-Gen}, an LLM-based Flow Matching framework that leverages STAR-VAE's structured latent space for high-fidelity generation without suffering from vector quantization artifacts. Empirical results demonstrate that STAR-VAE successfully resolves the trilemma, achieving state-of-the-art reconstruction fidelity and enhanced semantic information preservation across diverse audio domains. The structured latent space improves both traditional diffusion models and our \textbf{STAR-Gen} paradigm, achieving state-of-the-art performance in text-to-audio generation. The project page is available at~\url{https://STAR-VAE.github.io}.

Deep Learning · Generative Models and Autoencoders

Omar Bacarreza, Thorin Farnsworth, Alexander Makarovskiy, Hugo Wallner, Tessa Hicks, Santiago Sempere-Llagostera, John Price, Robert Francis-Jones, William Clements

Many successful families of generative models leverage a low-dimensional latent distribution that is mapped to a data distribution. Though simple latent distributions are often used, the choice of distribution has a strong impact on model performance. Recent experiments have suggested that the probability distributions produced by quantum processors, which are typically highly correlated and classically intractable, can lead to improved performance on some datasets. However, when and why latent distributions produced by quantum processors can improve performance, and whether these improvements are connected to quantum properties of these distributions, are open questions that we investigate in this work. We show in theory that, under certain conditions, these "quantum latent distributions" enable generative models to produce data distributions that classical latent distributions cannot efficiently produce. We provide intuition as to the underlying mechanisms that could explain a performance advantage on real datasets. Based on this, we perform extensive benchmarking on a synthetic quantum dataset and the QM9 molecular dataset, using both simulated and real photonic quantum processors. We find that the statistics arising from quantum interference lead to improved generative performance compared to classical baselines, suggesting that quantum processors can play a role in expanding the capabilities of deep generative models.

Deep Learning · Generative Models and Autoencoders

Dorian Gailhard, Enzo Tartaglione, Lirida Naviner, Jhony H. Giraldo

Graph generative models perform well on small structured data but struggle to scale to large, complex structures. Hierarchical approaches improve scalability but often ignore node and edge features, which are critical in real-world applications, particularly for hypergraphs that model higher-order relationships. In this paper, we propose FAHNES (feature-aware (hyper)graph generation via next-scale prediction), a hierarchical framework that jointly generates topology and features for graphs and hypergraphs. FAHNES builds multi-scale representations through node coarsening and localized expansion, guided by a novel hierarchical scale encoding that controls granularity and ensures cross-scale consistency. Experiments on synthetic, 3D mesh, and graph point cloud datasets demonstrate state-of-the-art performance in joint structure and feature generation.

Deep Learning · Generative Models and Autoencoders

Junhao Song, Yu Zhou, William J. Knottenbelt, Yudong Cao

The scaling hypothesis assumes that increasing model parameters yields emergent reasoning capabilities. This position paper argues that applying this probabilistic paradigm to generic quantum circuit synthesis is a category error. Unlike natural languages, quantum circuits require strict adherence to mathematical constraints, such as unitarity. Training on unverified code constitutes data poisoning. Models learn syntax but fail to capture the physical semantics of Hilbert space. Since the valid subset of circuit designs decays exponentially with the number of qubits, post-hoc filtering is mathematically intractable. We propose a pivot from human-centric copilots to verifier-centric agents. We integrate hierarchical constraints, topological masks, and symbolic proxies directly into generation. Our analysis suggests that scale alone cannot bridge the validity gap. Verification-aware architectures offer a viable path for modular quantum program generation. The community must stop simulating the physicist and instead satisfy the physical rules.

Applications · Everything Else

Youngsung Kim

Differentiable discrete selection uses soft mixtures during training but hard selection at deployment, resulting in a training-inference gap. We decompose this gap into selection gap (method-dependent, reducible) and computation gap (input-dependent, irreducible). Our key finding: the selection gap is determined by forward-pass structure, not backward-pass gradients. Methods using hard selection during training achieve zero selection gap by construction, while mixture methods exhibit gaps even with identical gradient estimators. This occurs because mixtures reward hedging across options, while deployment requires commitment to one. We propose CAGE (Confidence-Adaptive Gate Exploration), which addresses optimization entirely in the backward pass by adapting temperature based on selection confidence. We also identify a critical failure mode: Gumbel-ST suffers 40--50 percentage point accuracy collapse at low temperatures, which CAGE prevents. Experiments on logic gate networks validate the theory: hard selection achieves 98% accuracy with zero gap across all temperatures.

Deep Learning · Generative Models and Autoencoders

Rylan Schaeffer, Joshua Kazdan, Alvan Arulandu, Sanmi Koyejo

The proliferation of AI-generated content online has fueled concerns over \textit{model collapse}, a degradation in future generative models' performance when trained on synthetic data generated by earlier models. Industry leaders, premier research journals and popular science publications alike have prophesied catastrophic societal consequences stemming from model collapse. In this position piece, we contend this widespread narrative fundamentally misunderstands the scientific evidence. We highlight that research on model collapse actually encompasses eight distinct and at times conflicting definitions of model collapse, and argue that inconsistent terminology within and between papers has hindered building a comprehensive understanding of model collapse. To assess how significantly different interpretations of model collapse threaten future generative models, we posit what we believe are realistic conditions for studying model collapse and then conduct a rigorous assessment of the literature's methodologies through this lens. While we leave room for reasonable disagreement, our analysis of research studies, weighted by how faithfully each study matches real-world conditions, leads us to conclude that certain predicted claims of model collapse rely on assumptions and conditions that poorly match real-world conditions, and in fact several prominent collapse scenarios are readily avoidable. Altogether, this position paper argues that model collapse has been warped from a nuanced multifaceted consideration into an oversimplified threat, and that the evidence suggests specific harms more likely under society's current trajectory have received disproportionately less attention.

Deep Learning · Large Language Models

Haorui Wang, Parshin Shojaee, Kazem Meidani, Kunyang Sun, Jose Miguel Hernandez-Lobato, Teresa Head-Gordon, Jiajun He, Chandan Reddy, Chao Zhang, Yuanqi Du

Large language models are increasingly used to accelerate scientific discovery, especially in iteratively searching scientific hypotheses. Yet in many discovery settings the goal is not to identify a single ``best'' hypothesis: validation is noisy and expensive, multiple hypotheses can remain plausible, and scientists benefit from a set of high-quality but meaningfully diverse hypotheses that hedge against downstream uncertainty. Nevertheless, commonly used evolutionary search recipes tend to underemphasize this requirement, implicitly prioritizing optimization over exploration, and the resulting selection pressure during the search process leads to diversity collapse. Motivated by these limitations, we formulate hypothesis search as a sampling problem, where the objective is to efficiently produce diverse, high-quality hypotheses under fixed validation budget. Building on this perspective, we propose, EvoDiverse, an evolutionary framework inspired by the classical parallel tempering algorithm that searches hypotheses at multiple temperature levels and enables principled information exchange across temperatures to improve exploration without disrupting convergence. Across domains including molecular discovery, equation discovery, and algorithm discovery, our approach consistently improves both hypothesis quality and diversity under the same validation budget, and produces candidate sets that remain robust under more expensive downstream computational validations.

Deep Learning · Attention Mechanisms

Sakshi Choudhary, Aditya Chattopadhyay, Luca Zancato, Elvis Nunez, Matthew Trager, Wei Xia, Stefano Soatto

Language models struggle to generalize beyond the context lengths seen during pretraining, limiting performance on long-horizon reasoning and retrieval. Continued pretraining on long-context data can mitigate this limitation, but it is prohibitively expensive due to the quadratic scaling of Attention with sequence length. In practice, most tokens do not require Global Attention over the entire sequence and can rely on local context. Based on this insight, we propose L2A, a sequence modeling layer that enables token-wise long-term conditional memory access by deciding \textit{when} to invoke Global Attention. We evaluate L2A on Qwen 2.5 and Qwen 3 models, extending their effective context length from 32K to 128K tokens, where it matches standard long-context training within 1.5–3\% while skipping Global Attention for $\sim$80\% of tokens and outperforming prior baselines. We also design custom Triton kernels to efficiently realize this token-wise conditional attention on GPUs, achieving up to $\sim$2× improvements in training throughput and time-to-first-token over FlashAttention-2. Moreover, L2A enables post-training pruning of highly sparse Global Attention layers, reducing KV cache memory by up to 50\% with negligible performance loss.

Deep Learning · Algorithms

Sannyuya Liu, Ao Chen, Lin Liu, Ruxia Liang, Xiaoxuan Shen, Jianwen Sun

Symbolic Regression aims to discover interpretable mathematical expressions from data. Equation Learner (EQL) is a gradient-based method with strong fitting capability and expressive potential, yet it often activates redundant operators as model complexity grows, leading to over-complex expressions and unstable equation recovery. We analyze a gradient residual issue induced by operators that do not vanish at zero, which can prevent the ideal sparse expression from being a local optimum and bias training toward unnecessarily complex structures, making exact recovery nearly unattainable in practice. To address this, we propose EQL-Z, a structurally controllable symbolic regression framework. EQL-Z enforces zero-point constraints via zero-point consistent operator transformations to eliminate residual gradients on silent paths, and performs an incremental small-to-large structure search that grows depth/width from a compact seed under a complexity-penalized validation score. After selecting a compact structure, we optionally apply BFGS fine-tuning to refine coefficients. Experiments on synthetic and real-world datasets show that EQL-Z substantially improves exact equation recovery and in-/out-of-distribution generalization over vanilla EQL, achieving performance close to the best existing symbolic regression baselines. The code is available at https://anonymous.4open.science/r/EQL-Z-BE6C/.

Deep Learning · Algorithms

Benhao Huang, Zhengyang Geng, Zico Kolter

Reasoning is central to building intelligent systems that can solve unseen problems beyond training. Yet we still lack a principled understanding of what internal mechanism enables neural networks to generalize reasoning beyond memorized patterns. We hypothesize that generalizable reasoning emerges through learning task-conditioned attractors. Concretely, the model learns a latent dynamical system whose fixed points correspond to valid solutions. We term models that reason by converging to such task-conditioned fixed points *Equilibrium Reasoners (EqR)*. This attractor view elucidates when and how to scale test-time compute. Empirically, improvements from scaling test-time compute are tightly coupled with convergence to attractors. By shaping a more favorable attractor landscape and leveraging stochasticity, EqR improves convergence and scales reliably at test time. Our models scale along two axes: *depth* by running more solver steps, and *width* by aggregating stochastic trajectories from multiple random initializations. As we scale test-time compute by $8192\times$, with max effective layers surpassing 300,000 layers when unrolled, reasoning accuracy rises from 8\% to over 99\% on Sudoku-Extreme. We hope our attractor perspective sheds light on scalable reasoning through test-time computation.

Optimization · Zero-order and Black-box Optimization

Yukun Du, Haiyue Yu, Jiang Jiang, Shuaiwen Tang, Xiaotong Xie, Haobo Liu, Chongshuang Hu, Shengkun Chang

Existing Meta-Black-Box Optimization (MetaBBO) methods focus on how to search when controlling optimizers, but largely overlook where to search. We propose MetaSG-SAEA, a bi-level MetaBBO framework for expensive constrained multi-objective optimization problems (ECMOPs), in which a meta-policy provides search guidance to the low-level Surrogate-Assisted Evolutionary Algorithm (SAEA). To achieve this, we introduce Max–Min Constraint-Calibrated Inequality (MM-CCI), a compact, problem-agnostic region abstraction that maps heterogeneous constraint evaluations to an ordered scalar level; we further provide a theoretical analysis of its fundamental properties. Building on this region abstraction, we adopt diffusion-based population initialization to translate the meta-policy’s region-level guidance into solution-level priors for the SAEA. To make MetaSG-SAEA scalable, we construct an attention-based state representation across varying problem dimensions, population sizes, and numbers of objectives and constraints. Experimental results demonstrate that MetaSG-SAEA outperforms state-of-the-art baselines across diverse benchmarks and exhibits the ability to generalize across problem distributions.

Applications · Time Series

Daniel Durstewitz, Christoph Jürgen Hemmer, Florian Hess, Charlotte Ricarda Doll, Lukas Eisenmann

Time series (TS) modeling has come a long way from early statistical, mainly linear, approaches to the current trend in TS foundation models. With a lot of hype and industrial demand in this field, it is not always clear how much progress there really is. To advance TS forecasting and analysis to the next level, here we argue that the field needs a *dynamical systems (DS)* perspective. TS of observations from natural or engineered systems almost always originate from some underlying DS, and arguably access to its governing equations would yield theoretically optimal forecasts. This is the promise of *DS reconstruction (DSR)*, a class of ML/AI approaches that aim to infer *surrogate models* of the underlying DS from data. But models based on DS principles offer other profound advantages: Beyond short-term forecasts, they enable to predict the *long-term statistics* of an observed system, which in many practical scenarios may be the more relevant quantities. DS theory furthermore provides domain-independent *theoretical insight into mechanisms* underlying TS generation, and thereby will inform us, e.g., about upper bounds on performance of *any* TS model, generalization into unseen regimes as in tipping points, or potential control strategies. After reviewing some of the central concepts, methods, measures, and models in DS theory and DSR, we will discuss how insights from this field can advance TS modeling in crucial ways, enabling better forecasting with much lower computational and memory footprints. We conclude with a number of specific suggestions for translating insights from DSR into TS modeling.

Deep Learning · Algorithms

Runze Tian, Peng Kou

Deep learning has been widely regarded as a powerful tool for Koopman operator theory-based modeling, as it provides a promising architecture for data-driven learning of observable functions. To fully leverage this advantage, a well-designed training paradigm is required. However, the existing training paradigms typically either incur high optimization complexity or hinder effective end-to-end training, limiting modeling accuracy and training efficiency. To address this issue, we propose a differentiable quadratic programming (QP)-embedded deep Koopman framework (QPKO). In QPKO, a QP problem, which comprises a one-step accuracy-oriented objective function and a set of multi-step accuracy-oriented constraints, is formulated to introduce a mapping from observable functions to the global linear model. By doing so, the global linear model no longer needs to be treated as an independent trainable component, thereby effectively reducing optimization complexity. This QP-based mapping is implemented as a differentiable and computationally efficient module by leveraging OptNet (a differentiable QP layer), enabling effective end-to-end training. Experiments on four nonlinear dynamical systems show that QPKO achieves satisfactory improvements in modeling accuracy, training efficiency, and control performance.

Deep Learning · Algorithms

Renjie Li, Tong Sun, Yi Gao, Wei Dong

While edge GPUs are increasingly used for latency-critical DNN tasks, limited resources often fail to meet strict real-time (RT) requirements under concurrent workloads. Existing preemption and early-exit mechanisms often underutilize GPU resources through single-task queuing and sacrifice excessive accuracy during task bursts. To address this, we propose RTInfer, a novel system that enables concurrent RT task execution while balancing throughput and accuracy. RTInfer integrates an accuracy-calibrated lightweight variant co-optimization to generate efficient models, a memory-layout-aware scheduler to mitigate fragmentation during preemption, and an on-demand loading strategy to minimize host-to-GPU latency. Extensive evaluations demonstrate that RTInfer outperforms state-of-the-art methods by up to 98.2\% in deadline miss rate (DMR) and 58.0\% in accuracy.

Social Aspects · Everything Else

Ujwal Kumar, Alice Saito, Hershraj Niranjani, Rayan Yessou, Tan Phan Xuan

Constitutional AI has focused on single-model alignment using fixed principles. However, multi-agent systems create novel alignment challenges through emergent social dynamics. We present Constitutional Evolution, a framework for automatically discovering behavioral norms in multi-agent LLM systems. Using a grid-world simulation with survival pressure, we study the tension between individual and collective welfare, quantified via a Societal Stability Score $\mathcal{S} \in [0,1]$ that combines productivity, survival, and conflict metrics. Adversarial constitutions lead to societal collapse ($\mathcal{S}=0$), while vague prosocial principles (''be helpful, harmless, honest'') produce inconsistent coordination ($\mathcal{S}=0.249$). Even constitutions designed by Claude 4.5 Opus with explicit knowledge of the objective achieve only moderate performance ($\mathcal{S}=0.332$). Using LLM-driven genetic programming with multi-island evolution, we evolve constitutions maximizing social welfare without explicit guidance toward cooperation. The evolved constitution $\mathcal{C}^*$ achieves $\mathcal{S}=0.556\pm0.008$ (123\% higher than human-designed baselines, $N=10$), eliminates conflict, and discovers that minimizing communication (0.9\% vs 62.2\% social actions) outperforms verbose coordination. Our interpretable rules demonstrate that cooperative norms can be discovered rather than prescribed.

Social Aspects · Everything Else

Rashid Mushkani

This position paper argues that prompts used to deploy large language models (LLMs) in public-sector settings should be treated as governed artefacts rather than private, transient inputs. Prompts encode role instructions, decision framings, and value claims; prompt choice can materially shift outputs even when model weights and input records are held fixed. Existing governance tools, including model and dataset documentation, organisation-level policies, and post-training alignment, rarely make the local prompt collections used in deployment transparent, contestable, or auditable. We propose Prompt Commons: a versioned, community-maintained repository of prompt templates with provenance metadata, licensing, and moderation logs. Using a pilot dataset collected with community partners in a large North American city (443 human prompts; 3,317 after augmentation), we illustrate three governance states (open, curated, veto-enabled) and a negotiation-oriented ensemble method that aggregates stakeholder prompts into compromise recommendations. We close with falsifiable implications and an evaluation agenda for prompt-layer governance.

Xiyuan Wang, Muhan Zhang

Expressivity has been a major focus in the design of Graph Neural Networks (GNNs), yet a significant gap persists between theoretical universal expressivity and practical performance. While many expressive GNNs are efficient and achieve strong results, they often focus on specific graph properties and lack theoretical expressivity for general graph tasks. Conversely, theoretically universal-expressive models often suffer from high computational costs or poor generalization, limiting their real-world applicability. To bridge this gap, we introduce Equivariant Noise GNNs (ENGNNs), a framework that utilizes random noise features to enhance the expressivity of GNNs. Crucially, unlike prior methods that naively use noise, we enforce equivariance to nodewise noise transformations, such as orthogonal transformations. We prove that this property reduces the model's theoretical sample complexity, thereby improving generalization. Our framework simultaneously reaches theoretical universal expressivity, maintains the linear scalability of standard Message-Passing Neural Networks in practice, and achieves performance comparable to computationally expensive, high-expressivity models. Extensive experiments confirm strong performance across node, link, subgraph, and graph-level prediction tasks, demonstrating that the equivariant use of noise provides a powerful and practical pathway for building expressive GNNs. Our code is available at https://anonymous.4open.science/r/EquivNoiseGNN.