ICML 2026 论文列表

Reinforcement Learning (RL) is pivotal for enhancing Large Language Model (LLM) reasoning, yet mainstream algorithms such as GRPO and DAPO remain constrained by a coarse-grained credit assignment paradigm, where all tokens within the same response receive the identical reward. In this paper, we propose **Dynamic Entropy Weighting**, systematically define entropy-based weight ratios $\frac{H_{i,t}}{\sum_{k=1}^{n} H_{k,t}}$ and similar variants to redistribute rewards and get fine-grained rewards through two new algorithms: **Group Token Policy Optimization (GTPO)**, which assigns an entropy-weighted reward to each token and synthesizes token-specific advantage function to drive the model toward optimal path, and the analogous algorithm **Sequence-Level GRPO (GRPO-S)**, which admits a completely similar design at the sequence level. Unlike methods using entropy as mere regularization, GTPO and GRPO-S establish a new state-of-the-art on AIME and MATH 500, outperforming prior entropy-guided baselines and validating our weighting mechanism.

The rapid progress of generative models such as GANs and diffusion models has led to the widespread proliferation of AI-generated images, raising concerns about misinformation, privacy violations, and trust erosion in digital media. Although large-scale multimodal models like CLIP offer strong transferable representations for detecting synthetic content, fine-tuning them often induces catastrophic forgetting, which degrades pre-trained priors and limits cross-domain generalization. To address this issue, we propose the Distillation-guided Gradient Surgery Network (DGS-Net), a novel framework that preserves transferable pre-trained priors while suppressing task-irrelevant components. Specifically, we introduce a gradient-space decomposition that separates harmful and beneficial descent directions during optimization. By projecting task gradients onto the orthogonal complement of harmful directions and aligning with beneficial ones distilled from a frozen CLIP encoder, DGS-Net achieves unified optimization of prior preservation and irrelevant suppression. Extensive experiments on 50 generative models demonstrate that our method outperforms state-of-the-art approaches by an average margin of 6.6%, achieving superior detection performance and generalization across diverse generation techniques.

In performative learning, the data distribution reacts to the deployed model—for example, because strategic users adapt their features to game it—which creates a more complex dynamic than in classical supervised learning. One should thus not only optimize the model for the current data but also take into account that the model might steer the distribution in a new direction, without knowing the exact nature of the potential shift. We explore how regularization can help cope with performative effects by studying its impact in high-dimensional ridge regression. We show that, while performative effects worsen the test risk in the population setting, when moving to the over-parameterized regime where the number of features exceeds the number of samples, the optimal regularization in the presence of performativity helps reduce the variance in the estimated parameters, thereby improving performance. We show that the optimal regularization scales with the overall strength of the performative effect, making it possible to set the regularization in anticipation of this effect. We illustrate this finding through empirical evaluations of the optimal regularization parameter on both synthetic and real-world datasets.

Evaluating representation similarity is fundamental to representation learning. However, existing metrics suffer from significant limitations: they lack interpretability due to shifting baselines, lack robustness to outliers, and are computationally intractable for large datasets, forcing reliance on heuristic approximations. To address this, we develop an ordinal-similarity framework, instantiated by the Triplet (TSI) and Quadruplet (QSI) Similarity Indices, which measure alignment by quantifying the consistency of ordinal relationships. We theoretically demonstrate this formulation is inherently interpretable, robust to outliers, and computationally efficient. Finally, we establish a formal equivalence between TSI and local neighborhood alignment, measured by Mutual Nearest Neighbors. Empirically, we validate these properties and show that ordinal similarity offers a scalable approach to measuring alignment, enabling practitioners to better understand and design representations.

While deep learning has revolutionized financial forecasting through sophisticated architectures, the design of the supervision signal itself is rarely scrutinized. We challenge the canonical assumption that training labels must strictly mirror inference targets, uncovering the Label Horizon Paradox: the optimal supervision signal often deviates from the prediction goal, shifting across intermediate horizons governed by market dynamics. We theoretically ground this phenomenon in a dynamic signal-noise trade-off, demonstrating that generalization hinges on the competition between marginal signal realization and noise accumulation. To operationalize this insight, we propose a bi-level optimization framework that autonomously identifies the optimal proxy label within a single training run. Extensive experiments on large-scale financial datasets demonstrate consistent improvements over conventional baselines, thereby opening new avenues for label-centric research in financial forecasting.

While Large Language Model (LLM) agents demonstrate proficiency in static benchmarks, their deployment in real-world scenarios is hindered by the dynamic nature of user queries, tool sets, and interaction dynamics.To address this generalization gap, we formalize **OpenAgent** (Tool-Use Agent in Open-World), a problem setting characterized by distributional shifts across query, action, observation, and domain dimensions.To systematically diagnose its impact, we construct a controlled sandbox environment where we define fine-grained environmental shifts across a four-tier hierarchy, *Perception*, *Interaction*, *Reasoning*, and *Internalization*, and conduct a comprehensive series of experiments. Our analysis yields a series of key insights, demonstrating that agents trained via both Supervised Fine-Tuning (SFT) and Reinforcement Learning suffer from varying degrees of performance degradation when confronting open environmental shifts.Building on these insights, we propose Perturbation-Augmented Fine-Tuning, a disturbance-based intervention strategy for SFT that lays the foundation for enhancing agent robustness and utility in realistic environments. Our code will be released at: [https://github.com/LAMDA-NeSy/OpenAgent](https://github.com/LAMDA-NeSy/OpenAgent).

Cross-domain offline reinforcement learning leverages a source dataset to improve policy learning in a data-scarce target domain, but dynamics mismatch makes many source transitions kinematically infeasible and can cause negative transfer. Recent non-parametric geometric methods (e.g., standard optimal transport and k-nearest neighbors) avoid overfitting yet often yield only relative rankings under an implicit matching or retrieval budget, making performance sensitive to hand-tuned thresholds when the true cross-domain overlap is unknown. We formulate availability estimation as soft subset selection by learning a source reweighting that geometrically aligns with the target. We propose **R**obust **O**ffline unbalanced **O**ptimal **T**ransport (ROOT): (i) a robust ambiguity set for uncertainty under limited target samples, and (ii) an unbalanced transport objective that penalizes mass deviation, enabling a principled transport-or-discard mechanism. ROOT thus down-weights or discards high-cost source samples rather than forcing them onto the target support. Moreover, the induced weights decay exponentially with transport cost, guaranteeing outlier suppression. On D4RL dynamics-shift benchmarks, ROOT improves downstream offline RL and outperforms strong baselines on most tasks without task-specific threshold tuning.

Scaling test-time compute with multi-path chain-of-thought can improve reasoning accuracy, but its gains hinge on an effective exploration–exploitation trade-off. Existing methods handle this trade-off in rigid ways: tree-structured search hard-codes exploration via brittle expansion rules that disrupt post-trained reasoning, while parallel reasoning over-explores redundant hypothesis paths and relies on a weak answer selection strategy. Driven by the insight that the optimal balance is *phase-dependent* and that correct vs. incorrect paths often *diverge only at late stages*, we reconceptualize test-time scaling as a dynamic *expand–reduce* control problem over a pool of hypothesis paths. We introduce **HyPER**, a *training-free online control policy* for MoE multi-path decoding that reallocates compute under a fixed budget using lightweight path statistics. HyPER features (i) an *online controller* that shifts from exploration to exploitation as the hypothesis pool evolves, (ii) an MoE-based token-level refinement primitive for efficient *generation-time exploitation* without full-path resampling, and (iii) a length- and confidence-aware aggregation rule to bridge the existence–selection gap for reliable *answer-time exploitation*. Extensive experimental results across four MoE models and diverse benchmarks demonstrate HyPER consistently achieves the accuracy–compute Pareto frontier, outperforming prior-art methods by 8-10% while reducing token consumption by 25-40%.

Consider the problem of finding the best matching in a weighted graph where we only have access to predictions of the actual stochastic weights, based on an underlying context. If the predictor is the Bayes optimal one, then computing the best matching based on the predicted weights is optimal. However, in practice, this perfect information scenario is not realistic. Given an imperfect predictor, a suboptimal decision rule may compensate for the induced error and thus outperform the standard optimal rule. In this paper, we propose multicalibration as a way to address this problem. This fairness notion requires a predictor to be unbiased on each element of a family of protected sets of contexts. Given a class of matching algorithms $\mathcal C$ and any predictor $\gamma$ of the edge-weights, we show how to construct a specific multicalibrated predictor $\hat \gamma$, with the following property. Picking the best matching based on the output of $\hat \gamma$ is competitive with the best decision rule in $\mathcal C$ applied onto the original predictor $\gamma$. We complement this result by providing sample complexity bounds, and by performing numerical experiments.

This work proposes a framework LGKDE that learns kernel density estimation for graphs. The key challenge in graph density estimation lies in effectively capturing both structural patterns and semantic variations while maintaining theoretical guarantees. Combining graph kernels and kernel density estimation (KDE) is a standard approach to graph density estimation, but has unsatisfactory performance due to the handcrafted and fixed features of kernels. Our method LGKDE leverages graph neural networks to represent each graph as a discrete distribution and utilizes maximum mean discrepancy to learn the graph metric for multi-scale KDE, where all parameters are learned by maximizing the density of graphs relative to the density of their well-designed perturbed counterparts. The perturbations are conducted on both node features and graph spectra, which helps better characterize the boundary of normal density regions. Theoretically, we establish consistency and convergence guarantees for LGKDE, including bounds on the mean integrated squared error, robustness, and generalization. We validate LGKDE by demonstrating its effectiveness in recovering the underlying density of synthetic graph distributions and applying it to graph anomaly detection across diverse benchmark datasets. Extensive empirical evaluation shows that LGKDE demonstrates superior performance compared to state-of-the-art baselines on most benchmark datasets.

Pursuing training-free open-vocabulary semantic segmentation in an efficient and generalizable manner remains challenging due to the deep-seated spatial bias in CLIP. To overcome the limitations of existing solutions, this work moves beyond the CLIP-based paradigm and harnesses the recent spatially-aware dino.txt framework to facilitate more efficient and high-quality dense prediction. While dino.txt exhibits robust spatial awareness, we find that the semantic ambiguity of text queries gives rise to severe mismatch within its dense cross-modal interactions. To address this, we introduce VIsual-guided Prompt evolution (VIP) to rectify the semantic expressiveness of text queries in dino.txt, unleashing its potential for fine-grained object perception. Towards this end, VIP integrates alias expansion with a visual-guided distillation mechanism to mine valuable semantic cues, which are robustly aggregated in a saliency-aware manner to yield a high-fidelity prediction. Extensive evaluations demonstrate that VIP: I) surpasses the top-leading methods by 1.4%~8.4\% average mIoU, II) generalizes well to diverse challenging domains, and III) requires marginal inference time and memory overhead.

With the rise of reasoning language models and test-time scaling methods as a paradigm for improving model performance, substantial computation is often required to generate multiple candidate sequences from the same prompt. This enables exploration of different reasoning paths toward the correct solution, however, allocates the same compute budget for each prompt. Grounded on the assumption that different prompts carry different degrees of complexity, and thus different computation needs, we propose EAGer, a training-free generation method that leverages model uncertainty through token-wise entropy distribution to reduce redundant computation and concurrently improve overall performance. EAGer allows branching to multiple reasoning paths only in the presence of high-entropy tokens, and reallocates the saved compute budget to instances where exploration of alternative paths is most needed. We validate EAGer across multiple open-source models on complex reasoning benchmarks, with gains specifically demonstrated on AIME 2025. When target labels are accessible -- as in RLVR training pipelines -- EAGer achieves up to +37% in Pass@k and 59% fewer tokens; in test-time settings it still yields +12% in Pass@k and 64% fewer tokens compared to Full Parallel Sampling.

Most reinforcement learning (RL) baselines maximize future cumulative rewards with a fixed single discount factor, which limits their performance in complex sequential decision-making tasks due to a failure to balance short-term objectives and long-term planning. To address this issue, this paper focuses on a multi-timescale critic framework, where each component corresponds to a Q-value with a distinct discount factor. Two key improvements are proposed: (1) A Neural Reward Decoder reconstructs the reward sequence from multi-scale Q-values, with value and reward reconstruction losses enhancing Q-value estimation consistency; (2) A cross-attention-based Q-weight predictor adaptively adjusts Q-value weights via current observations to generate the final Q-value for policy optimization. Extensive experiments on DMControl and CARLA benchmarks demonstrate that our method significantly outperforms state-of-the-art (SOTA) baselines. Furthermore, we validate the framework's generalizability by integrating it with both off-policy (SAC, DrQ-v2) and on-policy (PPO) algorithms, achieving consistent performance gains. The code is available in the supplementary material.

AI predictive systems increasingly support high-stakes decision making, yet robust decisions under uncertainty often rely on human capabilities beyond AI alone. This motivates collaborative approaches that combine human judgment with AI predictions. We study this problem through the lens of uncertainty quantification and introduce **Human-AI Collaborative Uncertainty Quantification**, a framework in which an AI system refines a human expert’s proposed prediction set subject to two principles: **counterfactual harm**, requiring that the AI not degrade correct human judgments, and **complementarity**, requiring recovery of correct outcomes the human missed. At the population level, we show that the optimal collaborative prediction set has a simple two-threshold structure over a single score function, governing pruning and augmentation relative to the human proposal. Building on this characterization, we develop offline and online calibration algorithms with **distribution-free** finite-sample guarantees. The online algorithm adapts to arbitrary distribution shifts, including settings where human behavior evolves through interaction with the AI. Empirically, we show that collaborative prediction sets outperform human-only and AI-only baselines, achieving improved coverage--efficiency tradeoffs across image classification, regression, and text-based medical decision making.

Data plays an invaluable role in large-scale ML training pipelines. Multiple factors, including the need to incentivize the creation of high-quality data and efforts to compensate creative data work, have led to increased interest in data pricing. Data pricing mechanisms seek to establish a market where data providers are compensated based (in part) on the value of their data to the data buyer, e.g., frontier AI labs. However, assessing the exact value that each provider's data adds to the data buyer's objective requires repeated re-training, which is infeasible in practice. Our work studies data pricing under compute constraints. In our setting, data buyers cannot make data acquisition decisions optimally due to limited compute. Inspired by existing practice in the field of data selection, we propose a model for this problem called ``pricing with an attribution oracle,'' and provide a theoretical analysis of compute-efficient acquisition and pricing.

The integration of audio modality into Large Audio Language Models (LALMs) significantly expands their attack surface. Existing jailbreaks predominantly treat audio as a carrier for malicious payloads, relying on semantic optimization, acoustic parameter control, or additive perturbation to embed harmful content into the audio signal. In this work, we challenge this necessity and propose a new paradigm in which the role of audio shifts from content injection to safety alignment interference. We reveal that LALM safety alignment can be compromised solely by specific **Acoustic Latent Semantics (ALS)**, the underlying paralinguistic features intrinsic to the priors of audio generative models. Distinct from previous works that leverage explicit acoustic parameters to merely style malicious audio, we demonstrate that interference audio, benign in content but infused with specific ALS, can serve as a universal jailbreak trigger. Leveraging this insight, we propose **Acoustic Interference Attack (AIA)**, which decouples the attack payload from the audio. It employs a set of universal, instruction-neutral interference audio, enabling standard malicious text queries to bypass safety alignment without instance-specific optimization. Experiments on 10 LALMs across five datasets demonstrate that AIA achieves the state-of-the-art attack success rate. Furthermore, our interpretability analysis uncovers the inference path drift induced by AIA and identifies the inherent effective patterns within ALS, revealing the fundamental vulnerability of cross-modal alignment in LALMs.

The goal of single-channel source separation is to reconstruct $K$ sources given their mixture. In supervised settings where vast amounts of clean source data are available, this challenging, ill-posed problem has been addressed successfully by generative diffusion and flow-based prior models. However, access to such clean source samples is often limited, and even when available, supervised models are vulnerable to domain shifts. To bridge this gap, we present Separation via Unsupervised Remixing Flow (\textbf{SURF}), an unsupervised flow matching approach for source separation that learns directly from observed mixtures. This method relies on a novel combination of state-of-the-art supervised flow matching and regression-based self-supervised techniques. At a high level, starting from a teacher model, we utilize a ``remixing'' step to bootstrap the learning of a student flow model from the teacher's estimates. We provide insights into the objectives optimized by this approach and draw a novel connection to the Wake-Sleep algorithm. Empirical evaluations on image and audio benchmarks demonstrate that \textbf{SURF} establishes a new state-of-the-art, significantly outperforming existing unsupervised methods. See our demo page for examples. https://google.github.io/df-conformer/surf/

Self-correction is essential for solving complex reasoning problems in vision–language models (VLMs), yet existing reinforcement learning (RL) methods struggle to learn it. Effective self-correction behaviors emerge only rarely during RL, making learning signals sparse. To address this challenge, we propose c**o**rre**ct**i**o**n-s**p**ecific rollo**u**t**s**} (**Octopus**), a rollout-augmentation framework that synthesizes dense self-correction supervision by recombining existing rollouts without computational overhead. This rollout augmentation simultaneously improves sample efficiency and stabilizes RL optimization. Furthermore, we introduce a two-stage RL training strategy that disentangles self-correction and direct reasoning, avoiding signal conflicts and enabling both behaviors to be learned effectively. Building on this, we introduce $\texttt{Octopus-8B}$, an advanced reasoning VLM with controllable self-correction capabilities. It achieves SoTA performance among open-source VLMs across 7 benchmarks, outperforming the best RLVR baseline by 1.0 score while requiring only $0.72\times$ training time per step.

Current deep learning models for Multispectral and Hyperspectral Image Fusion (MS/HS fusion) are typically designed for fixed spectral bands and spatial scales, which limits their transferability across diverse sensors. To address this, we propose SSA, a unified framework for MS/HS fusion with spectral-band and fusion-scale agnosticism. Specifically, we introduce Matryoshka Kernel (MK), a novel operator that enables a single model to process varying ordered spectral band counts within a supported range. Meanwhile, we build SSA upon an Implicit Neural Representation (INR) backbone that models the HS signal as a continuous function, enabling reconstruction at arbitrary spatial resolutions. Together, these two mechanisms support a single MS/HS fusion model for heterogeneous sensors and arbitrary query scales. Extensive experiments demonstrate that our single jointly trained model achieves state-of-the-art performance, generalizes to unseen spatial scales, and transfers to unseen sensors with few-step adaptation. The code can be obtained at https://github.com/vg219/SSA.