ICML 2026 论文列表

LLM-augmented database analytics face a major bottleneck in the costly prefill phase. Although relational tables inherently contain repeated attribute values, standard row-by-row processing produces fragmented prompt layouts that obscure shared prefixes, thereby minimizing opportunities for prefix KV cache reuse and constraining system efficiency. Existing solutions typically employ heuristic or exhaustive search methods to reorder prompt layouts, but these approaches can be inefficient and may not leverage the structural properties of relational tables. We address this challenge by formulating prefix-cache-aware prompt layout optimization as a problem rooted in the isomorphism between prefix-cache reuse and the radix tree topology induced by the relational data distribution. Building on this perspective, we introduce a practical greedy tree-shaping algorithm that efficiently selects row and column orderings to maximize prefix overlap. Our approach, SOLO, improves prefill throughput by up to 90.3\% under fixed prefix-cache budget. Moreover, it reduces planning overhead by up to 242$\times$ compared to state-of-the-art baselines.

Large-scale AI evaluation increasingly relies on aggregating binary judgments from $K$ annotators, including LLMs used as judges. Most classical methods, e.g., Dawid-Skene or (weighted) majority voting, assume annotators are conditionally independent given the true label $Y\in\{0,1\}$, an assumption often violated by LLM judges due to shared data, architectures, prompts, and failure modes. Ignoring such dependencies can yield miscalibrated posteriors and even confidently incorrect predictions. We study label aggregation through a hierarchy of dependence-aware models based on Ising graphical models and latent factors. For class-dependent Ising models, the Bayes log-odds is generally quadratic in votes; for class-independent couplings, it reduces to a linear weighted vote with correlation-adjusted parameters. We present finite-$K$ examples showing that methods based on conditional independence can flip the Bayes label despite matching per-annotator marginals. We prove separation results demonstrating that these methods remain strictly suboptimal as the number of judges grows, incurring nonvanishing excess risk under latent factors. Finally, we evaluate the proposed method on three real-world datasets, demonstrating improved performance over the classical baselines.

LLM serving exhibits extreme length variability, making size-based scheduling difficult in practice. Recent LLM schedulers approximate SJF/SRPT using predicted decode lengths or rank and primarily report mean-centric metrics (e.g., TTFT/TBT). We show these prediction-driven policies can be fragile under distribution shifts, bursty arrivals, and GPU memory pressure, and still offer limited control over tail latency (P90–P99) that dominates user experience—even with perfect decode-length knowledge. We introduce a distribution-aware, prediction-free scheduling framework that replaces explicit length prediction with soft, $\gamma$-parameterized priority boosting driven by lightweight statistical signals. Our design co-optimizes scheduling with cache-aware preemption to account for memory-coupled decode dynamics that vary across workload mixes. Evaluated on Azure production traces, our method achieves a P99 TTLT up to 35--50\% lower than SRPT with perfect length prediction and a TTFT 34--47\% lower across various workloads, including reasoning-heavy and chat-heavy tasks, demonstrating a robust alternative for tail-latency optimization in online LLM serving.

Large language models have achieved impressive performance across diverse tasks. However, their tendency to produce overconfident and factually incorrect outputs, known as hallucinations, poses risks in real world applications. Conformal prediction provides finite-sample, distribution-free coverage guarantees, but standard conformal prediction breaks down under domain shift, often leading to under-coverage and unreliable prediction sets. We propose a new framework called Domain-Shift-Aware Conformal Prediction (DS-CP). Our framework adapts conformal prediction to large language models under domain shift, by systematically reweighting calibration samples based on their proximity to the test prompt, thereby preserving validity while enhancing adaptivity. Our theoretical analysis and experiments on the MMLU benchmark demonstrate that the proposed method delivers more reliable coverage than standard conformal prediction, especially under substantial distribution shifts, while maintaining efficiency. This provides a practical step toward trustworthy uncertainty quantification for large language models in real-world deployment.

Point clouds provide a compact and expressive representation of 3D objects, and have recently been integrated into multimodal large language models (MLLMs). However, existing methods primarily focus on static objects, while understanding dynamic point cloud sequences remains largely unexplored. This limitation is mainly caused by the lack of large-scale cross-modal datasets and the difficulty of modeling motions in spatio-temporal contexts. To bridge this gap, we present 4DPC$^2$hat, the *first* MLLM tailored for dynamic point cloud understanding. To this end, we construct a large-scale cross-modal dataset 4DPC$^2$hat-200K via a meticulous two-stage pipeline consisting of topology-consistent 4D point construction and two-level captioning. The dataset contains over 44K dynamic object sequences, 700K point cloud frames, and 200K curated question–answer (QA) pairs, supporting inquiries about counting, temporal relationship, action, spatial relationship, and appearance. At the core of the framework, we introduce a Mamba-enhanced temporal reasoning MLLM to capture long-range dependencies and dynamic patterns among a point cloud sequence. Furthermore, we propose a *failure-aware* bootstrapping learning strategy that iteratively identifies model deficiencies and generates targeted QA supervision to continuously strengthen corresponding reasoning capabilities. Extensive experiments demonstrate that our 4DPC$^2$hat significantly improves action understanding and temporal reasoning compared with existing models, establishing a strong foundation for 4D dynamic point cloud understanding.

To address the increasing long-context compute limitations of softmax attention, several subquadratic recurrent operators have been developed. This work includes models such as Mamba-2, DeltaNet, Gated DeltaNet (GDN), and Kimi Delta Attention (KDA). As the space of recurrences grows, a parallel line of work has arisen to taxonomize them. One compelling view is the test-time regression (TTR) framework, which interprets recurrences as performing online least squares updates that learn a linear map from the keys to values. Existing delta-rule recurrences can be seen as first-order approximations to this objective, but notably ignore the curvature of the least-squares loss during optimization. In this work, we address this by introducing preconditioning to these recurrences. Starting from the theory of online least squares, we derive equivalences between linear attention and the delta rule in the exactly preconditioned case. Next, we realize this theory in practice by proposing a diagonal approximation: this enables us to introduce preconditioned variants of DeltaNet, GDN, and KDA alongside efficient chunkwise parallel algorithms for computing them. Empirically, we find that our preconditioned delta-rule recurrences yield consistent performance improvements across synthetic recall benchmarks and language modeling at the 340M and 1B scale.

Proximal constraints are fundamental to the stability of the Large Language Model reinforcement learning. While the canonical clipping mechanism in PPO serves as an efficient surrogate for trust regions, we identify a critical bottleneck: fixed bounds strictly constrain the upward update margin of low-probability actions, disproportionately suppressing high-advantage tail strategies and inducing rapid entropy collapse. To address this, we introduce **Band-constrained Policy Optimization** (BandPO). BandPO replaces canonical clipping with **Band**, a unified theoretical operator that projects trust regions defined by $f$-divergences into dynamic, probability-aware clipping intervals. Theoretical analysis confirms that Band effectively resolves this exploration bottleneck. We formulate this mapping as a convex optimization problem, guaranteeing a globally optimal numerical solution while deriving closed-form solutions for specific divergences. Extensive experiments across diverse models and datasets demonstrate that BandPO consistently outperforms canonical clipping and Clip-Higher, while robustly mitigating entropy collapse. Code is publicly available at https://github.com/OpenMOSS/BandPO.

Affine Maximizer Auctions (AMAs), a generalized mechanism family from VCG, are widely used in automated mechanism design due to their inherent dominant-strategy incentive compatibility (DSIC) and individual rationality (IR). However, as the payment form is fixed, AMA's expressiveness is restricted, especially in distributions where bidders' valuations are correlated. In this paper, we propose Correlation-Aware AMA (CA-AMA), a novel framework that augments AMA with a new correlation-aware payment. We show that any CA-AMA preserves the DSIC property and formalize finding optimal CA-AMA as a constraint optimization problem subject to the IR constraint. Then, we theoretically characterize scenarios where classic AMAs can perform arbitrarily poorly compared to the optimal revenue, while the CA-AMA can reach the optimal revenue. For optimizing CA-AMA, we design a practical two-stage training algorithm. We derive that the target function's continuity and the generalization bound on the degree of deviation from strict IR. Finally, extensive experiments showcase that our algorithm can find an approximate optimal CA-AMA in various distributions with improved revenue and a low degree of violation of IR.

Spiking Neural Networks (SNNs) enable low-power, event-driven computation, but Transformer-based SNNs remain difficult to deploy on neuromorphic hardware due to dense operations and communication overhead. We propose two simplified attention mechanisms, \textbf{Simplified Spiking Attention (SSA)} and \textbf{Ultra-Simplified Spiking Attention (USSA)}, which replace matrix multiplication with Hadamard products and eliminate hardware-unfriendly components such as multi-head attention and scaling. We show that consecutive masking is redundant and analyze a spiking-order effect in which early spiking contributes more temporal information to attention modulation. On CIFAR-10, CIFAR-100, and DVS-Gesture, SSA achieves accuracies of 96.38\%, 79.45\%, and 97.56\%, respectively, while reducing computational complexity from $\mathcal{O}(N^2D)$ to $\mathcal{O}(ND)$ and communication complexity from $\mathcal{O}(N^2)$ to $\mathcal{O}(ND)$. USSA further reduces communication complexity to $\mathcal{O}(N)$ with only marginal accuracy degradation. On ImageNet-1K, SSA and USSA achieve 76.91\% and 77.27\% accuracy, respectively, demonstrating scalability to large-scale classification.

Block-sparse attention is promising for accelerating long-context LLM pre-filling, yet identifying relevant blocks efficiently remains a bottleneck. Existing methods typically employ coarse-grained attention as a proxy for block importance estimation, but often resort to expensive token-level searching or scoring, resulting in significant selection overhead. In this work, we trace the inaccuracy of standard coarse-grained attention via mean pooling to a theoretical root cause: the interaction between mean pooling and Rotary Positional Embeddings (RoPE). We prove that mean pooling acts as a low-pass filter that induces destructive interference in high-frequency dimensions, effectively creating a "blind spot" for local positional information (e.g., slash patterns). To address this, we introduce Prism, a training-free spectral-aware approach that decomposes block selection into high-frequency and low-frequency branches. By applying energy-based temperature calibration, Prism restores the attenuated positional signals directly from pooled representations, enabling block importance estimation using purely block-level operations, thereby improving efficiency. Extensive evaluations confirm that Prism maintains accuracy parity with full attention while delivering up to $\mathbf{5.1\times}$ speedup. Code available at https://github.com/xinghaow99/prism.

Hallucinations in large vision-language models (LVLMs) remain a critical challenge, where models often generate tokens that fail to align with visual evidence. To address this issue, we propose AFS: Anchor-Final Self-Supervision, a novel framework for hallucination-aware optimization in LVLMs. By leveraging discrepancies between intermediate and final layer predictions, AFS selectively applies self-supervision to visually descriptive tokens, incorporates hallucination-aware token classification, and encourages consistency between intermediate and final layer distributions. Unlike traditional methods that rely on explicit supervision or post-hoc interventions, AFS optimizes the model via Group Relative Policy Optimization (GRPO), using token-specific rewards derived from internal model signals. Experiments demonstrate that AFS significantly reduces hallucinations without compromising recall in caption generation. Beyond captioning, AFS excels in discriminative tasks, improving the reliability of object existence predictions and multimodal reasoning. Furthermore, AFS demonstrates strong cross-dataset generalization, transferring effectively across diverse visual domains. Code is available at \url{https://github.com/guavayew/AFS}.

Recommender systems (RS) may suffer from feedback-driven exposure concentration, where repeated engagement optimization collapses exposure onto a narrow set of categories, reducing catalog coverage and degrading long-horizon learning. Existing methods are often post hoc and typically lack principled uncertainty-aware risk estimates for regulating exposure under endogenous feedback. We therefore propose **CORAL**, a model-agnostic, uncertainty-aware framework that formulates exposure regulation as a constrained sequential decision problem. Specifically, we model self-reinforcing interactions to construct an exposure-saturation state, then derive an upper confidence bound on category-conditioned violation risk from observed history and incorporate it through a state-dependent penalty for adaptive intervention near saturation. Moreover, we provide theoretical guarantees for risk bounds, finite-time recovery, and efficient long-term performance. Extensive experiments on real-world datasets and controlled simulations validate the effectiveness of the proposed framework, which aligns with our theoretical analysis. Our code is available at: https://github.com/downw/CORAL.

While Mamba2's expanded state dimension enhances temporal modeling, it incurs substantial inference overhead that saturates bandwidth during autoregressive generation. Standard pruning methods fail to address this bottleneck: unstructured sparsity leaves activations dense, magnitude-based selection ignores runtime dynamics, and gradient-based methods impose prohibitive costs. We introduce GHOST (Grouped Hidden-state Output-aware Selection and Truncation), a structured pruning framework that approximates control-theoretic balanced truncation using only forward-pass statistics. By jointly measuring controllability and observability, GHOST rivals the fidelity of gradient-based methods without requiring backpropagation. As a highlight, on models ranging from 130M to 2.7B parameters, our approach achieves a 50% state-dimension reduction with approximately 1 perplexity point increase on WikiText-2. Code is available at https://github.com/Menezmic21/mamba2_ghost.

Block-wise diffusion language models (DLMs) generate multiple tokens in parallel, offering a promising alternative to autoregressive decoding. However, their inference efficiency remains bottlenecked by memory-bound attention in long-context scenarios. Naïve sparse attention is ineffective for DLMs due to the KV inflation problem: different queries select different prefix positions, causing the union of accessed KV pages to remain large. To address this challenge, we observe that block-wise diffusion exhibits locality of representation changes across denoising steps: only a small fraction of tokens (active tokens) undergo significant hidden-state updates, while most tokens (stable tokens) remain nearly unchanged. Based on this insight, we propose LoSA (Locality-aware Sparse Attention), which reuses cached prefix-attention results for stable tokens and applies sparse attention only to active tokens with large representation changes. This design reduces the number of queries contributing to the union of KV indices, substantially shrinking the KV pages that must be loaded. Across multiple block-wise DLMs and reasoning benchmarks, LoSA preserves near-dense accuracy while significantly improving efficiency, achieving up to 4.14× speedup over dense attention on RTX A6000 GPUs. LoSA also delivers up to 5% average improvement over baselines across all datasets and configurations, demonstrating the effectiveness of the proposed method.

Sparse Mixture-of-Experts (MoE) architectures scale model capacity efficiently but suffer from massive static parameter footprints, creating significant deployment burdens on memory-constrained hardware. Existing post-training pruning methods often rely on scalar statistics, ignoring the representational geometry of expert feature spaces. This leads to sub-optimal resource allocation across layers and the retention of redundant experts. To address this, we propose a Rank-aware Geometric Expert Pruning (RaGEP) framework to compress MoE models by analyzing the geometric properties of expert activations. First, in the inter-layer allocation stage, we introduce a Rank-aware budget allocation mechanism that adaptively assigns expert budgets based on the effective rank of layer-wise representations. Second, in the intra-layer selection stage, we propose a Spectral-Salience Pruning metric that harmonizes subspace orthogonality and activation magnitude to identify high-energy orthogonal experts. Extensive experiments across MoE models of different scales show that our method consistently outperforms state-of-the-art baselines on a diverse set of zero-shot tasks, while reducing model size and inference cost.

Cosine similarity estimation is a core primitive in coarse-to-fine retrieval pipelines, where early-stage candidate selection relies on approximate similarity estimates whose errors are amplified downstream. Widely used sign-based sketches arising from extreme quantization of random projections exhibit a structural variance peak near $\theta$ $\approx$ 90$^\circ$, the near-background region where candidate selection is most difficult. We propose QA-Cos, a query-aware decoder-side estimator that departs from the Hamming-agreement paradigm, treating sign bits as probabilistic observations rather than deterministic votes. Across simulations and BEIR benchmarks, QA-Cos reduces estimation error by up to $\sim$15--20\% in the near-orthogonal region and translates these gains into improved candidate selection in two-stage ANN pipelines, improving Hit@K by up to $\sim$30 percentage points at fixed budgets and reducing candidates by up to $\sim$45--50\% at fixed recall. In a native hnswlib scorer-replacement experiment, the same decoder improves the quality of the final HNSW frontier, and in a focused storage-aware setting a practical gated variant reduces end-to-end wall-clock latency at matched recall.

Federated Graph Learning (FGL) enables collaborative training across distributed clients without sharing raw graph data. However, its performance is severely hindered by graph-specific heterogeneity arising from divergent node feature distributions and disparate graph structures. Existing FGL methods primarily focus on aligning or personalizing node features but largely overlook the role of structural knowledge, leading to aggregation-induced representation drift during message passing. We observe that structural heterogeneity often originates from feature-driven connection biases shaped by local data collection practices or user preferences. To address this, we propose \textbf{Fed-Kalter}, a novel FGL framework that integrates Kalman filtering principles into graph neural networks. Fed-Kalter introduces Kalter-Conv, a graph convolution grounded in a Kalman framework with learnable dynamics, which treats structural embeddings as latent states and feature-augmented neighborhoods as noisy observations, thereby filtering feature-induced structural noise in a layer-wise manner. Only structural parameters are aggregated globally, enabling effective cross-client knowledge transfer while preserving local personalization. Extensive experiments on 16 graph classification datasets spanning 4 domains demonstrate that Fed-Kalter consistently outperforms state-of-the-art FGL methods. Further ablation and hyperparameter studies confirm its robustness, efficiency, and effectiveness in mitigating structural heterogeneity.

Recovering surface normals from single-view polarization images in complex media remains challenging. This paper focuses on ice as a representative complex medium, where intricate light–matter interactions lead to a nonlinear mapping between polarization observations and surface normals. To address this, a structure-aware polarization prior based on autocorrelation functions is proposed to capture the local spatial consistency of AoLP. Building on this, a dual-branch network (IceSfP) is designed to integrate raw polarization features with priors via cross-modal attention and multi-scale feature fusion, enabling accurate surface normal estimation under complex media conditions. To evaluate the method, the first real-world ice SfP dataset is constructed. Experimental results show that the method outperforms existing approaches across all metrics, achieving a MAE of 16.01$^\circ$, which is 2.74$^\circ$ lower than the second-best method. The framework provides a generalizable solution for high-precision geometric perception in complex media.

Early and accurate detection of anomalies in time-series data is critical due to the substantial risks associated with false or missed detections. While MLP-based mixer models have shown promise in time-series analysis, they do not maintain temporal causality during data processing. Moreover, real-world multivariate time series often contain numerous channels with diverse inter-channel correlations. Spurious correlations in the reconstructed time series lead to noisy representations, resulting in inaccurate anomaly detection. In addition, anomaly scoring methods that ignore temporal continuity can mislead sequential detection. To address these challenges, we propose a cluster-aware causal mixer for multivariate time-series anomaly detection. Channels are grouped into clusters based on their correlations, and each cluster is embedded through a dedicated embedding layer. A causal mixer is introduced to integrate information while maintaining temporal causality. We further develop a sequential anomaly-scoring method that accumulates evidence over time and refines anomaly boundaries. Our proposed model operates in an online fashion, making it suitable for real-time time-series anomaly detection. Experimental evaluations across six public benchmark datasets demonstrate that the proposed approach consistently achieves superior performance.