ICML 2026 Papers

Large Language Models (LLMs) have advanced Table Question Answering, where most queries can be answered by extracting information or simple aggregation. However, a common class of real-world queries is implicitly predictive, requiring the inference of unobserved answers from historical patterns rather than mere retrieval. These queries introduce two challenges: recognizing latent intent and reliable predictive reasoning over massive tables. To assess LLMs in such Tabular questiOn answering with implicit Prediction tasks, we introduce TopBench, a benchmark consisting of 779 samples across four sub-tasks, ranging from single-point prediction to decision making, treatment effect analysis, and complex filtering, requiring models to generate outputs spanning reasoning text and structured tables. We evaluate diverse models under both text-based and agentic workflows. Experiments reveal that current models often struggle with intent recognition, defaulting to just lookups. Deeper analysis identifies that accurate intent disambiguation serves as the prerequisite for leading these predictive behaviors. Furthermore, elevating the upper bound of prediction precision requires the integration of more sophisticated modeling or reasoning capabilities.

Handwritten Text Recognition (HTR) is computationally imbalanced in two ways: most image pixels are background, and many width-axis sequence positions are blank-dominated. This creates a mismatch for Spiking Neural Networks (SNNs): handwriting is observed as a static image, whereas spiking computation unfolds over timesteps. We propose Spike-HTR, a hybrid spiking recognizer that controls both the number of spiking steps and the number of width positions processed by the deep sequence mixer. To make a static image suitable for short-horizon spiking inference, InkCoder converts it into a coarse-to-fine input stream, where early steps cover broad stroke regions and later steps emphasize sharper stroke details. To reduce sequence computation, a CTC-guided length reducer keeps likely character or uncertain positions and compresses long blank-dominated stretches before deep mixing. With $T{=}2$, Spike-HTR trains only on target data, decodes without language models or lexicons, and reaches validation/test CERs of 3.5/5.4, 2.3/2.5, and 4.2/3.9 on IAM, LAM, and READ2016. Codes are available at https://github.com/QomolangmaH/SpikeHTR.

Most existing cultural evaluation frameworks for large language models (LLMs) focus on matching model outputs to ground-truth answers, primarily measuring factual cultural awareness. This overlooks whether models internalize broader cultural structure and pluralism. We introduce a spectral-analysis-based framework that captures large-scale macrostructural patterns in models' cultural knowledge and evaluate eight LLMs across nine cultural domains spanning all five of Newmark's cultural dimensions and 170 countries. Comparing with human data, we find that instruction-tuned models align more closely with human cultural structure than older models, while increased model size does not consistently improve performance. Finally, simulation-based experiments show that our proposed spectral metric better predicts a model's ability to serve users from unfamiliar cultural backgrounds than existing ones.

Evaluating alignment in language models requires testing how they behave under realistic pressure, not just what they claim they would do. While alignment failures increasingly cause real-world harm, comprehensive evaluation frameworks with realistic multi-turn scenarios remain lacking. We introduce an alignment benchmark spanning 904 scenarios across six categories---Honesty, Safety, Non-Manipulation, Robustness, Corrigibility, and Scheming---validated as realistic by human raters. Our scenarios place models under conflicting instructions, simulated tool access, and multi-turn escalation to reveal behavioral tendencies that single-turn evaluations miss. Evaluating 24 frontier models using LLM judges validated against human annotations, we find that even top-performing models exhibit gaps in specific categories, while the majority of models show consistent weaknesses across the board. Factor analysis reveals that alignment behaves as a unified construct (analogous to the g-factor in cognitive research) with models scoring high on one category tending to score high on others. We publicly release the benchmark and an interactive leaderboard to support ongoing evaluation, with plans to expand scenarios in areas where we observe persistent weaknesses and to add new models as they are released.

Pruning is a practical approach to compress large language models (LLMs), but it can amplify text degeneration, especially repetition loops, even when perplexity and task accuracy remain largely unchanged. In this work, we present a token-level analysis of this failure mode by viewing decoding as a dynamical process that enters and persists in a small set of recurrent contexts. Our analysis decomposes degeneration into loop entry risk and loop persistence, and shows that persistence is controlled by the escape mass assigned to plausible alternatives within the token sampling set. Motivated by these findings, we propose two token-level guidance objectives for post-pruning fine-tuning. FOCUS reweights distillation toward high-confidence teacher regions to suppress leakage, while RePAIR uses onset-centered positive/negative continuation pairs with a margin loss to promote plausible alternatives and prevent early commitment to repetition loops. Experiments on open-ended continuation and instruction-based generation show that both methods consistently reduce repetition and improve generation quality.

Standard reward models typically predict scalar scores that fail to capture the multifaceted nature of response quality in non-verifiable domains, such as creative writing or open-ended instruction following. To address this limitation, we propose Rubric-ARM, a framework that jointly optimizes a rubric generator and a judge using reinforcement learning from preference feedback. Unlike existing methods that rely on static rubrics or disjoint training pipelines, our approach treats rubric generation as a latent action learned to maximize judgment accuracy. We introduce an alternating optimization strategy to mitigate the non-stationarity of simultaneous updates, providing theoretical analysis that demonstrates how this schedule reduces gradient variance during training. Extensive experiments show that Rubric-ARM achieves state-of-the-art performance among baselines on multiple benchmarks and significantly improves downstream policy alignment in both offline and online reinforcement learning settings.

*How can we distinguish whether a peer review was written by a human or generated by an AI model?* We argue that, in this setting, authorship should not be attributed solely from the textual features of a review, but also from the ideas, judgments, and claims it expresses. To this end, we propose Sem-Detect, an authorship detection method for peer reviews that operationalizes this principle by combining textual features with claim-level semantic analysis. Sem-Detect compares a target review against multiple AI-generated reviews of the same paper, leveraging the observation that different AI models tend to converge on similar points, while human reviewers introduce more unique and diverse ones. As a result, Sem-Detect is able to distinguish fully AI-generated reviews from authentic human-written ones, including those that have been refined using an LLM but still reflect human judgment. Across a dataset of over 20,000 peer reviews from ICLR and NeurIPS conferences, Sem-Detect improves over the strongest prior detector by 36.5% in TPR@1% FPR in the binary setting. More importantly, in the three-class scenario, we empirically show that LLM refinement preserves the semantic signals of human reviews, which remain distinct from the patterns exhibited by fully AI-generated text; as a result, fewer than 3.5% of LLM-refined human reviews are misclassified as AI-generated.

AI-generated image (AIGI) detection is undergoing a critical transition from laboratory benchmarks to open-world adversarial defense. The prevalent paradigm focuses on finding static feature spaces, assuming that some invariant artifacts learned from historical data can achieve universal zero-shot generalization. While achieving saturation on several AIGI benchmarks, this static hypothesis suffers a severe performance drop against rapidly evolving generators (e.g., SD3, Nano Banana Pro). To address these limitations, we propose that the field should expand beyond "static generalization" to a new paradigm of "dynamic adaptation". We introduce **Fleet**, a framework that pioneers a dynamic paradigm of continuous few-shot evolution, enabling rapid alignment with emerging generative threats. Fleet improves few-shot adaptation by replacing unconstrained feature updates with constrained routing correction, where avoidance routing redirects novel AI samples away from Non-AI-dominated routes within decoupled subspaces. To validate this, we present **Treasure**, a benchmark spanning 64 models and 360k images, featuring diverse architectures and 20 closed-source commercial engines. Experiments reveal that while static SOTA methods fail catastrophically on modern generators, Fleet restores performance from 20.4\% to 73.1\% with only 10-shot adaptation on "Doubao Seedream 4.0". Code and data are available at https://github.com/ICTMCG/Fleet .

Large language models frequently exhibit hallucinations: fluent and confident outputs that are factually incorrect or unsupported by the input context. While recent hallucination detection methods have explored various features derived from attention maps, the underlying mechanisms they exploit remain poorly understood. In this work, we propose SinkProbe, a hallucination detection method grounded in the observation that hallucinations are deeply entangled with attention sinks - tokens that accumulate disproportionate attention mass during generation - indicating a transition from distributed, input-grounded attention to compressed, prior-dominated computation. Importantly, although sink scores are computed solely from attention maps, we find that the classifier preferentially relies on sinks whose associated value vectors have large norms. Moreover, we show that previous methods implicitly depend on attention sinks by establishing their mathematical relationship to sink scores. Our findings yield a novel hallucination detection method grounded in theory that produces state-of-the-art results across popular datasets and LLMs.

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

Stochastic Variance Reduced Gradient (SVRG) and its variants aim to speed-up training by using gradient corrections. Originally proposed over a decade ago, these methods have never been connected to any Bayesian method at a fundamental level. Here, we fill this gap and derive surprising new connections of SVRG to a recently proposed Bayesian method called `posterior correction'. Our main contribution is to show that SVRG can be recovered as a special case of posterior-correction over isotropic-Gaussian posteriors. Novel extensions of SVRG are automatically obtained by using more flexible exponential-family posteriors. We derive two new such extensions by using Gaussian families: a Newton-like variant with novel Hessian corrections, and an Adam-like extension that scales to large problems. Our work is the first to connect SVRG to Bayes and use it to speed-up training.

Even for the gradient descent (GD) method applied to neural network training, understanding its optimization dynamics, including convergence rate, iterate trajectories, function value oscillations, and especially its implicit acceleration, remains a challenging problem. We analyze nonlinear models with the logistic loss and show that the steps of GD reduce to those of generalized perceptron algorithms (Rosenblatt, 1958), providing a new perspective on the dynamics. This reduction yields significantly simpler algorithmic steps, which we analyze using classical linear algebra tools. Using these tools, we demonstrate on a minimalistic example that the nonlinearity in a two-layer model can provably yield a faster iteration complexity $\tilde{\mathcal{O}}(\sqrt{d})$ compared to $\Omega(d)$ achieved by linear models, where $d$ is the number of features. This helps explain the optimization dynamics and the implicit acceleration phenomenon observed in neural networks. The theoretical results are supported by extensive numerical experiments. We believe that this alternative view will further advance research on the optimization of neural networks.

Federated Prompt Learning (FPL) adapts Vision-Language Models to privacy-sensitive medical imaging, typically via a textual tuning paradigm that assumes the frozen visual encoder provides a discriminative feature geometry. We argue this assumption breaks down in medical settings, leading to two geometric pathologies: (1) Intra-client: Medical Manifold Collapse, where high morphological similarity reduces the effective rank of visual features; and (2) Inter-client: Medical Topological Misalignment, where heterogeneous acquisition protocols induce inconsistent geometry across clients. To address these, we propose FedMAP, which shifts the paradigm to Visual Manifold Anchoring. FedMAP utilizes an LLM-derived codebook as a client-invariant synchronization signal to restructure the visual space, via Manifold Semantic Anchoring (MSA) and Topology Structural Alignment (TSA) to enforce consistent inter-class relations. Experiments on FedISIC, FedCamelyon17, and a private ultrasound dataset show that FedMAP consistently outperforms state-of-the-art methods, especially in high-noise regimes where manifold collapse is most severe.

Deep protein structure predictors such as AlphaFold provide confidence estimates (e.g., pLDDT) that are not calibrated and degrade under distribution shifts across experimental modalities, temporal changes, and disordered regions. We introduce \textbf{CalPro}, a prior-aware evidential conformal framework for shift-robust uncertainty quantification. CalPro combines three components: (i) a geometric evidential head outputting Normal Inverse Gamma distributions via graph neural networks; (ii) a differentiable calibration surrogate that shapes representations during training, followed by split-conformal calibration for finite-sample coverage; and (iii) domain priors (disorder, flexibility) encoded as soft constraints on predicted uncertainty. Theoretically, we derive structure-aware \emph{sensitivity bounds} for coverage degradation under distribution shift using PAC-Bayesian control over ambiguity sets, quantifying how miscoverage increases with model complexity and shift magnitude. Empirically, CalPro achieves at most 5 percentage points coverage degradation across modalities compared to 15 to 25 points for baselines, reduces calibration error by 30\% to 50\%, and improves downstream docking success from 52\% to 75\% when filtering by uncertainty. The framework extends beyond proteins to structured regression tasks where priors encode local reliability.

LLMs have traditionally scaled along dense dimensions, where performance is coupled with near-linear increases in computational cost. While MoE decouples capacity from compute, it introduces large memory overhead and hardware efficiency challenges. To overcome these, we propose token-indexed parameters as a novel, orthogonal scaling axis that decouple model capacity from FLOPs. Specifically, we introduce ReToken and MoRT, which augment Transformer layers with modulation vectors retrieved from auxiliary embedding tables. These vectors modulate the backbone via lightweight, element-wise operations, incurring negligible FLOPs overhead. Extensive experiments on both dense and MoE backbones, spanning from 190M to 9.8B parameters, demonstrate that our approach consistently reduces validation loss and significantly improves downstream task performance (e.g., +7.3 on ARC-C, +6.3 on GSM8K). Rigorous isoFLOPs analysis further confirms that MoRT fundamentally shifts the quality–compute Pareto frontier, achieving comparable model quality with 35\% less compute relative to vanilla MoE architectures, and we validate that token-indexed parameters exhibit a predictable power-law scaling behavior. Moreover, our efficient implementation ensures that the overhead introduced by ReToken and MoRT remains marginal.

Many fields collect large-scale temporal data through repeated measurements (`trials’), where each trial is labeled with a set of metadata variables spanning several categories. For example, a trial in a neuroscience study may be linked to a value from category (a): task difficulty, and category (b): animal choice. A critical challenge in time-series analysis is to understand how these labels are encoded within the multi-trial observations, and disentangle the distinct effect of each label entry across categories. Here, we present MILCCI, a novel data-driven method that i) identifies the interpretable components underlying the data, ii) captures cross-trial variability, and iii) integrates label information to understand each category's representation within the data. MILCCI extends a sparse per-trial decomposition that leverages label similarities within each category to enable subtle, label-driven cross-trial adjustments in component compositions and to distinguish the contribution of each category. MILCCI also learns each component’s corresponding temporal trace, which evolves over time within each trial and varies flexibly across trials. We demonstrate MILCCI’s performance through both synthetic and real-world examples, including voting patterns, online page view trends, and neuronal recordings.

We present a constructive lower bound of $\Omega(1/\sqrt{\varepsilon})$ for Frank-Wolfe (FW) when both the objective and the constraint set are smooth and strongly convex, showing that the known uniform $\mathcal{O}(1/\sqrt{\varepsilon})$ guarantees in this regime are tight. It is known that under additional assumptions on the position of the optimizer, FW can converge linearly. However, it remained unclear whether strong convexity of the set can yield rates \emph{uniformly} faster than $\mathcal{O}(1/\sqrt{\varepsilon})$, i.e., irrespective of the position of the optimizer. To investigate this question, we focus on a simple yet representative problem class: minimizing a strongly convex quadratic over the Euclidean unit ball, with the optimizer on the boundary. We analyze the dynamics of FW for this problem in detail and develop a novel computational approach to construct worst-case FW trajectories, which is of independent interest. Guided by these constructions, we develop an analytical proof establishing the lower bound.

Symbolic expressions can effectively characterize and predict circuit behavior, but deriving them directly from circuit schematics is challenging. This process requires accurate visual-to-symbolic construction of circuit structure from images and correct multi-step symbolic derivation, both of which impose strict correctness requirements. This work proposes AutoVSR, an automated framework for visual-to-symbolic generation of circuit expressions using Vision Language Models (VLMs). By reconstructing circuit diagrams into an executable intermediate representation (Executable IR) and leveraging a symbolic solver for reasoning, AutoVSR significantly improves the accuracy of symbolic expression generation. AutoVSR introduces two key innovations: an IR construction method guided by component rule retrieval and verification-based feedback, and a symbolic solver implemented as a planning agent equipped with a symbolic tool library for reliable multi-step derivation. Compared with end-to-end VLM approaches and specialized methods on the main symbolic expression generation task, AutoVSR achieves accuracy improvements of 30.01--59.45% and 41.96--51.84%, respectively. Moreover, AutoVSR surpasses closed-source state-of-the-art VLMs in inference cost and computational efficiency. Code is available at https://github.com/LongfeiLi1/AutoVSR.

Empirically, option-based hierarchical reinforcement (HRL) learning often produces longer and more diverse options when a deliberation cost is charged at option boundaries. However, when options are executed for many steps under an approximate dynamics model, small model errors compound along the option, degrading the quality of the resulting plan. In this work, we introduce the commitment loss to formalize the tradeoff between deliberation cost and model error as a function of option duration. We characterize how optimal termination probabilities vary with this tradeoff under two model-error mechanisms. First, the model is learned from finite data via maximum-likelihood estimation, producing statistical error that interacts with option duration. Second, we consider an input-driven setting where an exogenous input is only observed at option boundaries and evolves unobserved between them, creating a drift-induced mismatch between planned and realized dynamics. In both cases, we solve for the optimal termination behavior as a function of deliberation cost and the error scale, clarifying the behavior of some popular HRL algorithms that approach the deliberation cost as a heuristic.