ICML 2026 论文列表

We introduce FrontierCS, a benchmark of 240 open-ended problems across diverse areas of computer science, designed and reviewed by experts, including CS PhDs and top-tier competitive programming participants and problem setters. Unlike existing benchmarks that focus on tasks with known optimal solutions, FrontierCS targets problems where the optimal solution is unknown, but the quality of a solution can be objectively evaluated. Models solve these tasks by implementing executable programs rather than outputting a direct answer. FrontierCS includes algorithmic problems, which are often NP-hard variants of competitive programming problems with objective partial scoring, and research problems with the same property. For each problem, we provide an expert reference solution and an automatic evaluator. Combining open-ended design, measurable progress, and expert curation, FrontierCS provides a benchmark at the frontier of computer-science difficulty. Empirically, we find that frontier reasoning models still lag far behind human experts, and that simply increasing reasoning budgets does not close this gap on open-ended challenges. Moreover, these models struggle to identify internal equivalence classes, and existing agentic frameworks also exhibit brittleness on such problems due to overfitting. FrontierCS thus offers a new lens into model capabilities on real frontier computer science problems.

Standard classification treats all errors equally, but in applications such as content moderation and medical screening, mistakes on clear-cut cases are more costly than errors on ambiguous ones. From a contextual bandit framework, we propose normalized excess cost (NEC), a metric that weighs classification errors by per-example costs and reduces to standard error rate when costs are uniform. Costs can derive from annotator vote margins, distance from decision thresholds, or confidence ratings. Across text, image, and tabular benchmarks, we find that NEC is often substantially lower than error rate—models with 5% error rate can achieve 1.8% NEC—revealing that most mistakes concentrate on ambiguous, low-cost examples. We also find that incorporating costs into training via loss weighting, sampling strategies, or regression yields inconsistent benefits. Our framework provides a practical methodology for deriving and evaluating instance-level misclassification costs, even if cost-sensitive training offers limited benefit.

With the advance of federated learning (FL) in privacy-sensitive domains, the need for efficient and robust training becomes increasingly urgent. Communication bottlenecks, heterogeneous client distributions, and fairness requirements make it essential to select the “right” data and features for model training. Yet existing FL research often addresses feature selection and data selection separately, ignoring their interplay in real-world high-dimensional and noisy datasets, leading to suboptimal performance. In this paper, we propose a unified framework for data and feature selection by formulating the problem as a generalized CUR decomposition problem. We introduce FedGCUR, a practical framework that integrates a federated column-pivoted QR (FedCPQR) decomposition routine with per-silo row selection. We prove that FedCPQR produces exactly the same decomposition results as centralized CPQR and establish an upper bound of the reconstruction error of FedGCUR. Experiments on tabular tasks and vision benchmarks show strong downstream accuracy and absolute CUR reconstruction quality compared with representative data and feature selection baselines.

Riemannian diffusion models generalize score-based generative modeling to manifold-supported data via stochastic diffusion equations on the manifold. However, training requires sampling from and differentiating the manifold heat kernel, which is rarely available in closed form beyond a few highly symmetric manifolds. We propose a general approach that approximates the heat kernel by directly solving the manifold heat equation with a physics-informed neural network (PINN). Given an explicit manifold specification, we choose a coordinate system, derive the corresponding heat (Fokker--Planck) equation and a short-time asymptotic approximation, and then train a PINN to learn the log heat kernel. The resulting surrogate enables both forward noising (heat-kernel sampling) and conditional-score evaluation for denoising score matching. We demonstrate the method on diverse manifolds including $S^2$, $SO(3)$, $\mathrm{SPD}(n)$, and permutation-quotiented point clouds.

Transforming a large language model (LLM) into a vision-language model (VLM) can be achieved by mapping the visual tokens from a vision encoder into the embedding space of an LLM. Intriguingly, this mapping can be as simple as a shallow MLP transformation. To understand why LLMs can so readily process visual tokens, we need interpretability methods that reveal what is encoded in the visual token representations at *every* layer of LLM processing. In this work, we introduce LatentLens, a novel approach for mapping latent representations to descriptions in natural language. LatentLens encodes a large text corpus and stores contextualized token representations for each token in that corpus. Visual token representations are then compared to these contextualized representations and the top-$k$ nearest neighbor representations serve as descriptions of the visual token. We evaluate this method on 15 different VLMs, showing that commonly used methods, such as LogitLens, substantially underestimate the interpretability of visual tokens. With LatentLens instead, the majority of visual tokens are interpretable across all studied models and all layers. Qualitatively, we show that the descriptions produced by LatentLens are semantically meaningful and provide more fine-grained interpretations for humans compared to individual tokens. More broadly, our findings contribute new evidence on the alignment between vision and language representations and open up new directions for analyzing the latent representations of LLMs.

The state-transition (decay) matrix governs how fixed-size memory is updated and used, making it a core design in linear attention models. Prior work exploits rank-1 approximations to reduce the cost of constructing decay matrices, but this low-rank constraint also limits the expressive capacity. We therefore formulate decay-matrix design as an open optimization problem: maximizing expressiveness while introducing minimal additional cost. Inspired by the multi-head mechanism, we propose Head-in-Head, which introduces an additional mask matrix to structure memory partitioning and interactions within a single linear-attention head. This simple, generic, and efficient design: 1) enables a rank-$r$ approximation of the decay matrix with only a few extra parameters and 2) strengthens intra-head information interaction. We further develop mask normalization and a chunk-wise parallelization scheme to support efficient parallel training. Extensive experiments on synthetic benchmarks and language modeling tasks, together with visual analyses, show that Head-in-Head consistently improves baseline performance by enriching information diversity and strengthening intra-head interactions. Code available at: \url{https://github.com/msj-19/Head-in-Head-Linear-Attention}

Standard approaches to goal-conditioned reinforcement learning (GCRL) that rely on temporal-difference learning can be unstable and sample-inefficient due to bootstrapping. While recent work has explored contrastive and supervised formulations to improve stability, we present a probabilistic alternative, called survival value learning (SVL), that reframes GCRL as a survival learning problem by modeling the distribution of time-to-goal from each state. This structured distributional Monte Carlo perspective yields a closed-form identity that expresses the goal-conditioned value function as a discounted sum of survival probabilities, enabling value estimation via a hazard model trained via maximum likelihood on both event and right-censored trajectories. We introduce three practical value estimators, including finite-horizon truncation and two binned infinite-horizon approximations to capture long-horizon objectives. Experiments on offline GCRL benchmarks show that SVL combined with hierarchical actors matches or surpasses strong hierarchical TD and Monte Carlo baselines, excelling on complex, long-horizon tasks.

Searching for effective policies is notoriously challenging in high-dimensional musculoskeletal systems, where multiple muscles actuate individual joints. Although this redundancy complicates naive policy search, it also implies that effective control can be captured by a low-dimensional action manifold. To identify such a manifold, we introduce Joint-Space Empowerment (JoSE), a novel information-theoretic objective that quantifies how much control an agent has over its mechanical degrees-of-freedom. We frame manifold discovery as an optimal precoding problem—where a state-dependent precoder maps low-dimensional latent actions to high-dimensional muscle commands—and derive the optimal precoder in closed form under control-affine Gaussian dynamics. We show that manipulation policies learned on this manifold display significantly enhanced dexterity, sample efficiency, and improved generalization. These results present optimal precoding as a general information-theoretic paradigm for coordinating high-dimensional actuators to control low-dimensional features.

Vibe coding is a new software development paradigm in which human engineers prompt a large language model (LLM) agent to complete complex coding tasks with little supervision. Although vibe coding is increasingly adopted, is the generated code really safe to deploy in production? To investigate this question, we propose SUSVIBES, a benchmark consisting of 186 feature-request software engineering tasks from real-world open-source projects, for which human programmers committed vulnerable implementations. We evaluate 12 widely used coding agentic settings with frontier models on the benchmark. Disturbingly, all agents perform poorly in terms of software security. Although 57% of the solutions from SWE-Agent with Claude 4 Sonnet are functionally correct, only 11.8% are secure. Further experiments demonstrate that preliminary security strategies, such as augmenting the feature request with vulnerability hints, cannot mitigate these security issues. Our findings raise serious concerns about the widespread adoption of vibe coding, particularly in security-sensitive applications. The code and dataset are available at https://github.com/LeiLiLab/susvibes. The leaderboard is at https://leililab.github.io/ susvibes-leaderboard.

Autonomous code agents built on large language models are reshaping software and AI development through tool use, long-horizon reasoning, and self-directed interaction. However, this autonomy introduces a previously unrecognized security risk: agentic interaction fundamentally expands the LLM attack surface, enabling systematic probing and recovery of hidden system prompts that guide model behavior. We identify system prompt extraction as an emergent vulnerability intrinsic to code agents and present JustAsk, a self-evolving framework that autonomously discovers effective extraction strategies through interaction alone. Unlike prior prompt-engineering or dataset-based attacks, JustAsk requires no handcrafted prompts, labeled supervision, or privileged access beyond standard user interaction. It formulates extraction as an online exploration problem, using Upper Confidence Bound-based strategy selection and a hierarchical skill space spanning atomic probes and high-level orchestration. These skills exploit imperfect system-instruction generalization and inherent tensions between helpfulness and safety. Evaluated on 41 black-box commercial models across multiple providers, JustAsk consistently achieves full or near-complete system prompt recovery, revealing recurring design- and architecture-level vulnerabilities. Our results expose system prompts as a critical yet largely unprotected attack surface in modern agent systems.

Fine-tuning large language models (LLMs) on resource-constrained clients remains a challenging problem. Recent works have fused low-rank adaptation (LoRA) techniques with federated fine-tuning to mitigate challenges associated with client model sizes and data scarcity. Still, the heterogeneity of resources remains a critical bottleneck: while higher-rank modules generally enhance performance, varying client capabilities constrain LoRA's feasible rank range. Existing approaches attempting to resolve this issue either lack analytical justification or impose additional computational overhead, leaving a wide gap for efficient and theoretically-grounded solutions. To address these challenges, we propose federated sketching LoRA (FSLoRA), which leverages a sketching mechanism to enable clients to selectively update submatrices of global LoRA modules maintained by the server. By adjusting the sketching ratios, which determine the ranks of the submatrices on the clients, FSLoRA flexibly adapts to client-specific communication and computational constraints. We provide a rigorous convergence analysis of FSLoRA that characterizes how the sketching ratios affect the convergence rate. Through extensive experiments, we demonstrate that FSLoRA outperforms baselines and significantly improves training efficiency while preserving stable convergence.

Functional ANOVA offers a principled framework for interpretability by decomposing a model’s prediction into main effects and higher-order interactions. For independent features, this decomposition is well-defined, strongly linked with SHAP values, and serves as a cornerstone of additive explainability. However, the lack of an explicit closed-form expression for general dependent distributions has forced practitioners to rely on costly sampling-based approximations. We completely resolve this limitation for categorical inputs. By bridging functional analysis with the extension of discrete Fourier analysis, we derive a closed-form decomposition without any assumption. Our formulation is computationally very efficient. It seamlessly recovers the classical independent case and extends to arbitrary dependence structures, including distributions with non-rectangular support. Furthermore, leveraging the intrinsic link between SHAP and ANOVA under independence, our framework yields a natural generalization of SHAP values for the general categorical setting.

While Large Language Models (LLMs) have demonstrated strong zero-shot reasoning capabilities, their deployment as embodied agents still faces fundamental challenges in long-horizon planning. Unlike open-ended text generation, embodied agents must decompose high-level intents into actionable sub-goals while adhering to the constraints of a dynamic environment. Standard LLM planners frequently fail to maintain strategy coherence over extended horizons due to context window limitations or hallucinate state transitions that violate environment constraints. We propose GiG, a planning framework that structures embodied agents’ memory using a Graph-in-Graph architecture. Our approach employs a Graph Neural Network (GNN) to encode environmental states into embeddings, organizing these embeddings into action-connected execution trace graphs within an experience memory bank. GiG enables retrieval of structurally-similar priors, allowing agents to ground current decisions in relevant past structural patterns. Furthermore, we introduce a bounded lookahead module that leverages symbolic transition logic to enhance the agent’s planning capabilities through grounded action projections. We evaluate our framework on three embodied planning benchmarks—Robotouille Synchronous, Robotouille Asynchronous, and ALFWorld. Our method outperforms state-of-the-art baselines, achieving Pass@1 performance gains of up to 22% on Robotouille Synchronous, 37% on Asynchronous, and 15% on ALFWorld while maintaining comparable or lower computational cost.

Thinking Large Language Models (LLMs) used as judges for pairwise preferences remain noisy at the single-sample level, and common aggregation rules (majority vote, soft self-consistency, or instruction-based self-aggregation) are inconsistent when ties are allowed. We study inference-time compute (ITC) for evaluators that generate $n$ independent thinking--rating samples per item, and propose a principled, distribution-calibrated aggregation scheme. Our method models three-way preferences with a Bradley–Terry-Davidson formulation on rating counts, leveraging both polarity (margin among non-ties) and decisiveness (non-tie rate) to distinguish narrow margins from strong consensus. Across various evaluation benchmarks, our approach consistently reduces MAE and increases pairwise accuracy versus standard baselines, and when evaluated against human-consensus meta-labels, matches or exceeds individual human raters. These results show that carefully allocating ITC and aggregating with distribution-aware methods turns noisy individual model judgments into reliable ratings for evaluation.

We investigate the origins of massive activations in large language models (LLMs) and identify a specific layer named the Massive Emergence Layer (ME Layer), that is consistently observed across model families, where massive activations first emerge and subsequently propagate to deeper layers through residual connections. We show that, within the ME Layer both the RMSNorm and the FFN parameters jointly contribute to the emergence of massive activations. Once formed, the massive activation token representation remains largely invariant across layers, reducing the diversity of hidden representations passed to the attention module. Motivated by this limitation, we propose a simple and effective method to reduce the rigidity of the massive activation token. Our approach consistently improves LLM performance across multiple tasks, including instruction following and math reasoning, in both training free and fine tuning settings. Moreover, we show that our method mitigates attention sinks by selectively weakening their influence, elucidating their origin at the hidden state level and shedding new light on principled mitigation strategies.

Continual learning, enabling models to acquire new skills and knowledge without degrading existing capabilities, remains a fundamental challenge for foundation models. While on-policy reinforcement learning can reduce forgetting, it requires explicit reward functions that are often unavailable. Learning from expert demonstrations, the primary alternative, is dominated by supervised fine-tuning (SFT), which is inherently off-policy. We introduce Self-Distillation Fine-Tuning (SDFT), a simple method that enables on-policy learning directly from demonstrations. SDFT leverages in-context learning by using a demonstration-conditioned model as its own teacher, generating on-policy training signals that preserve prior capabilities while acquiring new skills. Across skill learning and knowledge acquisition tasks, SDFT consistently outperforms SFT, achieving higher new-task accuracy while substantially reducing catastrophic forgetting. In sequential learning experiments, SDFT enables a single model to accumulate multiple skills over time without performance regression, establishing on-policy distillation as a practical path to continual learning from demonstrations.

Kronecker adapters have emerged as a promising approach for fine-tuning large-scale models, enabling high-rank updates through tunable component structures. However, existing work largely treats the component structure as a fixed or heuristic design choice, leaving the dimensions and number of Kronecker components underexplored. In this paper, we identify component structure as a key factor governing the capacity of Kronecker adapters. We perform a fine-grained analysis of both the dimensions and number of Kronecker components. In particular, we show that the alignment between Kronecker adapters and full fine-tuning depends on component configurations. Guided by these insights, we propose Component Designed Kronecker Adapters (CDKA). We further provide parameter-budget–aware configuration guidelines and a tailored training stabilization strategy for practical deployment. Experiments across various architectures and modalities demonstrate the effectiveness of CDKA. Code is available at \url{https://github.com/rainstonee/CDKA}.

Execution-aware LLM agents offer a promising paradigm for learning from tool feedback, but such feedback can be expensive and slow to obtain, making online reinforcement learning (RL) less practical in certain scenarios. High-coverage hardware verification exemplifies this challenge due to its reliance on industrial simulators and non-differentiable execution signals. We propose LLM4Cov, an offline agent-learning framework that models verification as single-step state transitions guided by deterministic evaluators. Building on this formulation, we introduce execution-validated data curation, policy-aware agentic data synthesis, and worst-state-prioritized sampling to enable scalable learning under execution constraints. We further curate a reality-aligned benchmark adapted from an existing verification suite through a revised evaluation protocol. Using the proposed pipeline, a compact 4B-parameter model achieves 69.2\% pass rate and 90.4\% average coverage in CVDP-ECov under agentic evaluation, outperforming its teacher by 5.3\% and 10.5\%, demonstrating competitive performance against models an order of magnitude larger.

Low-rank tensor decomposition (TD) is usually effective on clean, fully observed data, but it often degrades under severe missingness or noise. Low-rankness is itself a useful but limited structural prior, and additional handcrafted priors (e.g., sparsity or smoothness) still fall short of capturing the rich statistics of real-world data. To compensate for this weak inductive bias under heavy corruption, one would like to inject a learned, data-driven prior; however, the state-of-the-art diffusion models are not readily compatible with current TD and tractable posterior inference. To address these challenges, we introduce DiffBCP, a hybrid-prior Bayesian CP decomposition framework that couples a cumulative shrinkage process prior over the CP factors for automatic rank selection with an off-the-shelf pre-trained diffusion model as an implicit data prior on the reconstructed tensor. To make posterior inference tractable despite the coupling among the likelihood, low-rank constraint, and diffusion prior, we develop a split Gibbs sampler: CP factors admit conjugate updates, while the diffusion block is sampled via low-rank-guided denoising. A noise-adaptive coupling schedule further reduces sensitivity to hand-tuned annealing. Experiments on image inpainting and denoising, including high-resolution out-of-distribution images, show consistent gains over Bayesian, nonlinear, and plug-and-play TD baselines.