ICML 2026 论文列表

Existing machine unlearning methods primarily focus on removing data influence after training completes, which is effective for many scenarios, but a complementary capability is needed when removal requests arise during ongoing training. We propose LMCleaner, an efficient and certified *online* unlearning framework that can process unlearning requests at any training step without waiting for training completion. Our key insight is that influence propagation can be decomposed into a trust region where linear approximation is accurate, and a residual that concentrates in a low-dimensional subspace and can be efficiently masked by calibrated noise. Building on this insight, we design an influence propagation truncation mechanism that treats mini-batch influence as atomic units, computes influence within a truncation window for efficient removal, and injects subspace-aware noise for certified privacy. Our theoretical analysis proves that the truncation residual decays exponentially with window size and that the unlearned model is $(\varepsilon, \delta)$-indistinguishable from retraining. Experiments demonstrate that LMCleaner achieves over $100\times$ computational savings compared to baselines while maintaining model utility and defending against membership inference attacks.

Large Vision-Language Models (LVLMs) exhibit sophisticated reasoning but remain susceptible to object hallucination. Deviating from the prevailing *attention intensity assumption*, we reveal a deeper dynamic structural misalignment: hallucination is triggered at decision-critical steps where specific attention heads, acting as risky mediators, decouple from visual evidence to lock onto language priors. This establishes a pathological shortcut that bypasses visual grounding. To dismantle this, we propose **Fox** (*F*aithfulness and *O*bservational-flow via e*X*pression-rectification), a training-free inference-time framework. **Fox** diagnoses structural misalignment using a visual attention entropy probe to localize risky mediators unsupervisedly. We then execute a targeted causal intervention via numerical logit saturation to physically sever the shortcut path. Finally, a conflict-gated cooperative decoding strategy reconciles interventional faithfulness with observational fluency. Extensive experiments demonstrate that **Fox** achieves SOTA performance, outperforming SID by $29.1\%$ while preserving linguistic richness. Code is available at <https://github.com/Cc2021start/Fox>.

We study the problem of recovering the parameters of a multivariate Ornstein–Uhlenbeck (OU) process from steady-state observational and interventional data. In many applications, such as large-scale gene perturbation experiments, only stationary “snapshot” measurements are available, making standard stochastic differential equation estimation methods that rely on time-series trajectories inapplicable. We first establish an identifiability result: one intervention per strongly connected component (SCC) of the drift graph suffices to recover all OU process parameters generically up to a global scaling factor. This holds provided that the SCC condensation graph is connected with a single root and certain spectral nondegeneracy assumptions hold. We propose a recursive learning algorithm that orders SCCs topologically and, for each component, isolates its marginal dynamics and solves a linear system derived from the steady-state moment equations, leveraging parameters recovered for upstream components. Building on this theoretical foundation, we propose a regularized least-squares estimator that jointly minimizes residuals of the steady-state mean and covariance equations across observational and interventional data. Experiments on synthetic and real datasets demonstrate the effectiveness of our method in recovering parameters and predicting unseen interventions.

Long-context language modeling is commonly framed as a scalability challenge of token-level attention, yet local-to-global information structuring remains largely implicit in existing approaches. Drawing on cognitive theories of discourse comprehension, we propose HiCI (Hierarchical Construction--Integration), a hierarchical attention module that constructs segment-level representations, integrates them into a shared global context, and broadcasts both to condition segment-level attention. We validate HiCI through parameter-efficient adaptation of LLaMA-2 (7B and 13B) and Qwen3-8B with only $\sim$4--5\% additional parameters, extending context to 100K/64K tokens for LLaMA-2-7B/13B and to 48K tokens for Qwen3-8B. Across language modeling, retrieval, and instruction-following benchmarks, HiCI yields consistent improvements over strong baselines, including matching proprietary models on topic retrieval and surpassing GPT-3.5-Turbo-16K on code comprehension. These results demonstrate the effectiveness of explicit hierarchical structuring as an inductive bias for long-context modeling.

Effectively retrieving, reasoning, and understanding multimodal information remains a critical challenge for agentic systems. Traditional Retrieval-augmented Generation (RAG) methods rely on linear interaction histories, which struggle to handle long-context tasks, especially those involving information-sparse yet token-heavy visual data in iterative reasoning scenarios. To bridge this gap, we introduce VimRAG, a framework tailored for multimodal Retrieval-augmented Reasoning across text, images, and videos. Inspired by our systematic study, we model the reasoning process as a dynamic directed acyclic graph that structures the agent states and retrieved multimodal evidence. Building upon this structured memory, we introduce a Graph-Modulated Visual Memory Encoding mechanism, with which the significance of memory nodes is evaluated via their topological position, allowing the model to dynamically allocate high-resolution tokens to pivotal evidence while compressing or discarding trivial clues. To implement this paradigm, we propose a Graph-Guided Policy Optimization strategy. This strategy disentangles step-wise validity from trajectory-level rewards by pruning memory nodes associated with redundant actions, thereby facilitating fine-grained credit assignment. Extensive experiments demonstrate that VimRAG consistently achieves state-of-the-art performance on diverse multimodal RAG benchmarks.

This paper proposes asymmetric perturbation, where only one player's payoff function is perturbed, for solving bilinear saddle-point optimization problems, commonly arising in minimax problems, game theory, and constrained optimization. Symmetric perturbation is known to require decreasing its strength to ensure convergence to a solution, i.e., an equilibrium in the original game, resulting in a slower rate. First, with asymmetric perturbation, we show that, for a sufficiently small perturbation strength, the equilibrium strategy of the asymmetrically perturbed game coincides with an equilibrium strategy of the original unperturbed game. Second, building on this coincidence, we construct a learning algorithm with a linear last-iterate convergence rate. Third, motivated by the fact that the coincidence relies on the perturbation strength being sufficiently small, we also provide a parameter-free variant, retaining the linear rate. Finally, we empirically demonstrate fast convergence toward equilibria in both normal-form and extensive-form games.

Fine-tuning large language models often undermines their safety alignment, a problem further amplified by harmful fine-tuning attacks in which adversarial data removes safeguards and induces unsafe behaviors. We propose SPARD, a defense framework that integrates Safety-Projected Alternating optimization with Relevance-Diversity aware data selection. SPARD employs SPAG, which optimizes alternatively between utility updates and explicit safety projections with a set of safe data to enforce safety constraints. To curate safe data, we introduce a Relevance–Diversity Determinantal Point Process to select compact safe data, balancing task relevance and safety coverage. Experiments on GSM8K and OpenBookQA under four harmful fine-tuning attacks demonstrate that SPARD consistently achieves the lowest average attack success rates, substantially outperforming state-of-the-art defense methods, while maintaining high task accuracy. Code is available at https://github.com/shuhao02/SPARD.

Physics obeys strict symmetries like rotational equivariance. However, the standard Transformer architectures widely used in physics foundation models do not enforce these constraints by construction. We introduce ReViT, a rotationally equivariant Vision Transformer framework for neural PDE solvers operating on grid-based physical fields that achieves exact equivariance for the discrete groups $C_4$ (2D) and the chiral octahedral group $O$ (3D), with bounded approximate $\mathrm{SO}(d)$ equivariance for continuous rotations. ReViT maps scalar and vector inputs into locally invariant representations derived from physics-based canonical bases, enabling the use of standard self-attention without symmetry violations. Built on a hierarchical Swin-style backbone with a precomputed reference basis pyramid, ReViT preserves equivariance across multi-scale operations. We evaluate ReViT on a wide range of 2D and 3D PDE benchmarks, such as Magnetohydrodynamics and Turbulent Channel Flows, demonstrating significant gains over state-of-the-art baselines. ReViT exhibits strong generalization, and reduces MSE by up to 65\% compared with the best-performing alternatives.

Recent advances have empowered large language models (LLMs) with remarkable fine-grained instruction-following capabilities in text generation tasks. However, embedding methods typically rely solely on the hidden state of the input's last token, limiting their ability to capture complete semantic signals distributed across the full output tokens. Moreover, existing discrete-to-continuous re-encoding approaches introduce semantic discontinuity. To address these limitations, we propose $\textbf{InstEmb}$, a novel instruction following embedding framework. InstEmb jointly optimizes two key aspects: (1) Input-Intrinsic semantic information, achieved by employing contrastive learning focused on the representation of the last input token, and (2) Output-Aware semantic information, captured through representation self-distillation leveraging learnable look-ahead tokens without introducing additional decoding latency. Additionally, we introduce $\textbf{Dual-Anchor Alignment Pooling (DAAP)}$, explicitly aligned with our dual training objectives. Extensive experiments demonstrate that InstEmb achieves state-of-the-art performance across multiple instruction following benchmarks without benchmark-specific supervised data.

Large-scale dedicated application of LLMs in diverse scenarios increasingly demands specialized model inference behavior under strict constraints of accuracy, latency, and memory. However, the heterogeneous and long-tailed nature of real-world specialized scenarios makes it difficult to obtain training data and optimize models. We study a practical inference-time specialization setting: given an LLM base, we compile a reusable, budget-bounded pathway/subnetwork within a specific scenario. Our approach is motivated by an empirical coupling phenomenon: input scenario sets aligned with similar representation subspaces (e.g., domain) in embedding space tend to activate a consistent and sparse set of internal reasoning pathways in model parameter space. To build the bridge between them, we propose probe-based SubspacePath Pruner with two core components: (1) Domain-Basis Synthesis (DBS) constructs a quasi-orthogonal basis of domain axes in embedding space, serving as a stable coordinate system. (2) Probe-based Scenario Pruning (PSP) uses efficient layer-wise linear probes to estimate axis alignment and compute budgeted head-wise pathways for a specific scenario. Experiments on LLaMA-2-13B show 29.3 average Recall on cross-domain tests (vs. 24.7 dense) and 21.6 on cross-dataset tests (vs. 25.5 dense) with 1.27x speedup at ~30% pruning ratio.

Real-world time series inherently exhibit complex temporal patterns. Within chaotic systems, significant mixing and entanglement occur between different time-varying modes. Given that time series exhibit distinctly different patterns at various sampling scales, downsampling to extract multiscale features is a common approach. However, conventional downsampling causes high-frequency components in the original signal, those exceeding the new Nyquist frequency, to undergo spectral folding. This erroneously introduces spurious low-frequency patterns, perceived as low-frequency noise, thereby leading to the ***aliasing problem***. To address this problem, we propose a Decomposition-Prevention-Fusion architecture framework called **DMANet**, which introduces the **D**ynamic **M**ultiscale **A**nti-Aliasing **Net**work. Specifically, DMANet comprises two key components: Multiscale Convolutional Downsampling, designed to capture temporal dependencies and inter-channel interactions, and an Anti-Aliasing Operation, which includes Pre-Sampling Anti-Aliasing Filtering and Post-Sampling Interpolation. These designs guarantee the fidelity of multiscale features before and after downsampling. We show that by mitigating the risk of aliasing, our proposed simple convolutional downsampling architecture achieves performance competitive with common baselines. Our codes are available at https://github.com/HenryZhou2024/DMANet.

Multi-view learning has been widely applied for sleep stage classification using multi-modal data. However, existing methods typically assume that different modalities are well-aligned, which is often unattainable in real-world scenarios, thereby compromising the reliability of the staging results. In this paper, we propose ConfSleepNet, a conflict-aware evidential framework that dynamically resolves inter-view conflicts. The framework consists of multi-view evidence extraction and conflict-aware aggregation. In the first phase, it learns category-related evidence from different modalities, which represents the degree of support for individual sleep stages. Considering the inherent characteristics of varying modalities, we propose hybrid category structures for different modalities to promote more reasonable evidence learning. In the second phase, view-specific opinions, including prediction results and uncertainty, are constructed from the learned evidence. Notably, we propose a novel conflict-aware aggregation method that integrates these view-specific opinions into a reliable joint decision. This mechanism can effectively resolve conflicts among opinions and synthesize them into a reliable joint decision. Both theoretical analysis and experimental results demonstrate the effectiveness of ConfSleepNet in sleep staging tasks. The code is available at https://github.com/By4te/ConfSleepNet_ICML2026/.

Temporal straightening describes how, across layers in autoregressive LLMs, the trajectory traced by the token representations of an input sequence becomes straighter, potentially enabling next-token prediction by linear extrapolation. However, a direct link between this trajectory and token-level behavior has been missing. We provide such a link by relating contextual curvature—a geometric measure of how sharply the representational trajectory bends over recent context—to next-token entropy. Across two models (GPT-2 XL and Pythia-2.8B), contextual curvature is correlated with entropy, and this relationship emerges during training. Perturbation experiments reveal selective dependence: manipulating curvature through trajectory-aligned interventions reliably modulates entropy, while geometrically misaligned perturbations have no effect. Finally, regularizing representations to be straighter during training modestly reduces token-level entropy without degrading validation loss. These results identify trajectory curvature as a task-aligned representational feature that influences behavioral uncertainty in LLMs.

In streaming Reinforcement Learning (RL), transitions are observed and discarded immediately after a single update. While this minimizes resource usage for on-device applications, it makes agents notoriously sample-inefficient, since value-based losses alone struggle to extract meaningful representations from transient data. We propose extending Self-Predictive Representations (SPR) to the streaming pipeline to maximize the utility of every observed frame. However, due to the highly correlated samples induced by the streaming regime, naively applying this auxiliary loss results in training instabilities. Thus, we introduce orthogonal gradient updates relative to the momentum target and resolve gradient conflicts arising from streaming-specific optimizers. Validated across the Atari, MinAtar, and Octax suites, our approach systematically outperforms existing streaming baselines. Latent-space analysis, including t-SNE visualizations and effective-rank measurements, confirms that our method learns significantly richer representations, bridging the performance gap caused by the absence of a replay buffer, while remaining efficient enough to train on just a few CPU cores.

Diffusion Transformers (DiTs) have emerged as a powerful backbone for image generation, offering superior scalability over U-Nets. However, their practical deployment is hindered by significant computational costs. While Quantization-Aware Training (QAT) shows promise, its application to DiTs is challenged by the high sensitivity and complex distributions of activations. Identifying activation quantization as the primary bottleneck for low-bit settings, we propose **RobuQ**, a systematic QAT framework. We first establish a strong ternary weight (W1.58A4) baseline. Building on this, we introduce **RobustQuantizer**, which utilizes the Hadamard transform to convert unknown per-token distributions into normal distributions. Furthermore, we propose **AMPN**, the first **A**ctivation-only **M**ixed-**P**recision **N**etwork pipeline, applying ternary weights globally while allocating layer-specific activation precisions to eliminate information bottlenecks. Extensive experiments demonstrate that **RobuQ** achieves state-of-the-art performance on **ImageNet-1K**, representing the first stable image generation with activations quantized to an average of 2 bits. Code is available at https://github.com/racoonykc/RobuQ.

Zero-shot classifier expansion aims to recognize unseen classes by learning a shared mechanism to map semantics of all classes to classifier weights without access to images. However, existing methods rely on a shared mapping, which is difficult to classify in scenarios containing a mixture of distinct and similar classes, especially with the continuous expansion of classes. Since this mapping prioritizes general attributes for distinct classes while neglecting subtle attributes for similar ones, this granularity mismatch, compounded by sensitivity to noise, induces optimization interference where gradients from distinct classes dominate the learning process. To overcome this limitation, a granularity-aware adaptive framework with interventions is introduced to balance them. Specifically, this method first generates multi-source semantics by intervening on non-causal noise, then discovers latent class structure to separate distinct classes, and finally refine similar classes to synthesize weights with invariance to non-causal noise. The effectiveness is demonstrated through theoretical and empirical analysis in multiple aspects.

Learning rate configuration is a fundamental aspect of modern deep learning. The prevailing practice of applying a uniform learning rate across all layers overlooks the structural heterogeneity of Transformers, potentially limiting their effectiveness as the backbone of Large Language Models (LLMs). In this paper, we introduce Layerwise Learning Rate (LLR), an adaptive scheme that assigns distinct learning rates to individual Transformer layers. Our method is grounded in Heavy-Tailed Self-Regularization (HT-SR) theory, which characterizes the empirical spectral density (ESD) of weight correlation matrices to quantify heavy-tailedness. Layers with weaker heavy-tailedness are assigned larger learning rates to accelerate their training, while layers with stronger heavy-tailedness receive smaller learning rates. By tailoring learning rates in this manner, LLR promotes balanced training across layers, leading to faster convergence and improved generalization. Extensive experiments across architectures (from LLaMa to GPT-nano), optimizers (AdamW and Muon), and parameter scales (60M–3B, up to 100B tokens) demonstrate that LLR achieves up to 1.5× training speedup and outperforms baselines, notably raising average zero-shot accuracy from 47.09% to 49.02% for 1B models and from 48.58% to 50.61% for 3B models. A key advantage of LLR is its low tuning overhead: it transfers nearly optimal LR settings directly from the uniform baseline. Code is available at https://github.com/hed-ucas/Layer-wise-Learning-Rate.

We study the robustification of next-token prediction models in adversarial online decision-making environments. Suppose a next-token predictor is trained on a distribution over sequences of opponent actions, and a decision maker acts by approximately best responding to the model’s predictions. If this induced decision-making algorithm achieves low adversarial regret, we call such distributions low-regret distributions. For models with unbounded context, we show that every distribution is exponentially close in total variation distance to a low-regret distribution. Thus, sublinear regret can always be achieved while changing the original next-token prediction distribution by only a negligible amount. By contrast, for models with bounded context windows, as in standard transformer architectures, we prove an impossibility result: some distributions are $\Theta(1)$-far from every low-regret distribution. We complement these results with two positive findings. First, allowing a modest increase in context length restores robustification with near-optimal regret guarantees. Second, the unbounded-context robustification procedure can be implemented by adding a constant number of layers to a standard transformer architecture. Finally, we provide empirical evidence that small transformer models can be efficiently trained to represent these robustified low-regret distributions.

Machine unlearning aims to remove the influence of specific forget training data due to privacy, copyright or bias concerns while maintaining the model performance on the remaining retain data. Existing unlearning algorithms, such as optimizing a weighted combination of losses, have tried to achieve these objectives of improving forget quality and maintaining retain utility. However, they do not guarantee that these objectives can be improved by a specified extent for all forget and retain data. In this work, we address this limitation with a novel and theoretically-grounded approach from a constrained optimization perspective. Firstly, we identify that the *hardness* of reconciling both objectives can be quantified by the similarity between the forget data and the retain data. Next, we derive an unlearning algorithm (HAMU) with the overall goal of guaranteeing a specified improvement in forget quality while minimizing the retain utility cost/degradation by updating the model weights based on our hardness measure. Our hardness measure also informs users when retain utility degradation is unavoidable, i.e., both objectives cannot be improved simultaneously, and stopping should be considered. Our algorithm is applicable to non-convex models and is easily parallelizable, making it readily deployable in real-world scenarios. We empirically demonstrate HAMU's superior performance over baselines on both image and text datasets using large models. Our code is available at https://github.com/aoi3142/HAMU.