NeurIPS 2026 Papers

Post-training methods for large language models (LLMs) implicitly aggregate diverse human feedback in ways that can be studied explicitly using social choice theory. In this paper, we adopt the perspective that for each post-training method, there is a corresponding reference-dependent probabilistic voting rule. Taking Direct Preference Optimization (DPO) and Nash Learning from Human Feedback (NLHF) as our case studies, we study axiomatic properties of the associated voting rules. First, we show that NLHF violates one of the central axioms of social choice, namely monotonicity, if and only if its KL regularization parameter is below a bound, whereas DPO satisfies monotonicity regardless of its regularization parameter. Second, we show that while DPO is often associated with Borda-style preference aggregation, when DPO is regarded as a probabilistic voting rule, it violates two variable population axioms that the standard probabilistic version of Borda satisfies. Third, we show that unlike DPO, NLHF violates a commutativity axiom from Bayesian epistemology, which has practical implications for staged post-training. Finally, we conduct an empirical study of the frequency and magnitude of axiom violations in the CoVal dataset using real base models.

Central to human-aligned AI is understanding the benefits of human-elicited labels over synthetic alternatives. While human soft-labels improve calibration by capturing uncertainty, prior studies conflate these benefits with the implicit correction of mislabeled data (mode shifts), obscuring true effects of soft-labels. We present a controlled audit of soft-label learning across MNIST and a synthetic variant, re-annotating subsets to extract human uncertainty. By decoupling soft-label supervision from underlying label mode shifts, we show that while human soft-labels do provide accuracy gains, their larger value lies in acting as a regularizer that improves model calibration on difficult samples and promotes stable convergence across training runs. Dataset cartography reveals models trained on human soft-labels mirror human uncertainty, whereas those trained on synthetic labels fail to align with humans. Broadly, this work provides a diagnostic testbed for human-AI uncertainty alignment.

We propose a novel amortized optimization method for predicting optimal transport (OT) plans across multiple pairs of measures by leveraging Kantorovich potentials derived from sliced OT. We introduce two amortization strategies: regression-based amortization (RA-OT) and objective-based amortization (OA-OT). In RA-OT, we formulate a functional regression model that treats Kantorovich potentials from the original OT problem as responses and those obtained from sliced OT as predictors, and estimate these models via least-squares methods. In OA-OT, we estimate the parameters of the functional model by optimizing the Kantorovich dual objective. In both approaches, the predicted OT plan is subsequently recovered from the estimated potentials. As amortized OT methods, both RA-OT and OA-OT enable efficient solutions to repeated OT problems across different measure pairs by reusing information learned from prior instances to rapidly approximate new solutions. Moreover, by exploiting the structure provided by sliced OT, the proposed models are more parsimonious, independent of specific structures of the measures, such as the number of atoms in the discrete case, while achieving high accuracy. We demonstrate the effectiveness of our approaches on tasks including MNIST digit transport, color transfer, supply-demand transportation on spherical data, and mini-batch OT conditional flow matching.

Extracting signals through alpha factor mining is a fundamental challenge in quantitative finance. Existing automated methods primarily follow two paradigms: Decoupled Factor Generation, which treats factor discovery as isolated events, and Iterative Factor Evolution, which focuses on local parent-child refinements. However, both paradigms lack a global structural view, often treating factor pools as unstructured collections or isolated single-lineage paths, which leads to redundant search and inefficient allocation of optimization effort. To address these limitations, we introduce AlphaPROBE, (**Alpha** Mining via **P**rincipled **R**etrieval and **O**n-graph **B**iased **E**volution), a framework that reframes alpha mining as the strategic navigation of a Directed Acyclic Graph (DAG). By modeling factors as nodes and evolutionary links as edges, AlphaPROBE treats the factor pool as a dynamic, interconnected ecosystem. The framework consists of two core components: a Bayesian Factor Retriever that identifies high-potential seeds by balancing exploitation and exploration through a posterior probability model, and a DAG-aware Factor Generator that leverages the full ancestral trace of factors to produce context-aware, non-redundant optimizations. Extensive experiments on major Chinese and US stock market datasets against 8 competitive baselines demonstrate that AlphaPROBE significantly gains enhanced performance in predictive accuracy, return stability and training efficiency. Our results confirm that leveraging global evolutionary topology is essential for efficient and robust automated alpha discovery. We have open-sourced our implementation at https://anonymous.4open.science/r/AlphaPROBE-A103.

Language model benchmarks are pervasive and computationally-efficient proxies for real-world downstream performance. However, many recent works find that benchmarks often fail to predict downstream utility. While some works have begun diagnosing sources of misalignment, there remain no ways to systematically update benchmarks to align their scores with downstream usage. Towards bridging this gap, we introduce and study \textit{benchmark alignment}, where we use information about downstream model performance to automatically update benchmarks, aiming to produce new static benchmarks that predict model pairwise rankings on downstream tasks. Our experiments involving 4576 language models and 6 benchmarks show that benchmark items can successfully be reweighted to predict downstream performance for unseen models, even generalizing across model scales in most cases. And while naive alignment unsurprisingly requires large numbers of models and benchmark questions, we find that proper use of downstream data can often also enable alignment, even using responses from as few as 20 models. Overall, our work takes a step towards efficiently aligning benchmark development with downstream tasks.

Modern agricultural data is sourced from diverse platforms and spans multiple spatial scales, ranging from ground-level close-up photography to Unmanned Aerial Vehicle (UAV) aerial observation and satellite remote sensing imagery. Accordingly, agricultural multimodal reasoning demands robust cross-scale spatial understanding. However, due to the lack of multi-view agricultural benchmark datasets, existing multimodal large language models (MLLMs) exhibit severe ground-level bias, which leads to scale confusion then semantic collapse in agricultural perception tasks, such as misinterpreting farmland imagery as walls or floors. To address this, we introduce \textbf{AgroOmni}, a large-scale multi-view training corpus with 288K Visual Question Answering pairs covering 56 specialized task categories across 14 task types, designed to capture diverse scales in modern precision agriculture. Built on this dataset, we propose AgroNVILA, which achieves a new state-of-the-art of 62.32\% on the AgroMind benchmark ($+15.03\%$ over GPT-5.2), effectively mitigating the multi-view cross-scale gap for holistic agricultural understanding. Diagnostic evaluations on AgMMU further reveal an inherent heterogeneity between macro-priors and micro-diagnostics through constrained zero-shot performance. Meanwhile, even minimal fine-tuning leads to a dramatic performance gain of AgroNVILA on AgMMU, strongly demonstrating its generalization capability empowered by AgroOmni. Full training scripts are publicly available at \url{https://anonymous.4open.science/r/AgroOmni-6510}.

Recent AI systems achieve strong results on many benchmarks, yet these gains have not translated into economically meaningful deployment across professional domains. We argue that this gap is largely an evaluation problem: existing benchmarks rarely measure sustained performance on real, economically valuable workflows. We introduce Agents’ Last Exam (ALE), a benchmark for evaluating generalist computer-use AI agents on long-horizon professional tasks with verifiable outcomes. ALE is developed in collaboration with 250+ industry experts and organized around a taxonomy grounded in O*NET / SOC 2018. The benchmark spans 55 subdomains across 13 industry clusters and currently contains 1,490 task instances sourced from authentic professional workflows. Tasks require integrated GUI interaction, shell execution, software operation, and long-horizon planning inside real computing environments. To support scalable evaluation, ALE uses deterministic deliverable-based scoring and structured rubric verification rather than open-ended human judgment whenever possible. Experimental results show that current frontier agents remain far from saturation: across mainstream harness and backbone configurations, the average full-pass rate on the hardest tier is only 2.6%. More broadly, ALE is intended not merely as another leaderboard, but as an instrument for measuring the gap between benchmark success and GDP-relevant impact.

Acquiring high-quality labeled data and engineering effective learning pipelines are two critical bottlenecks to applying machine learning systems in new domains. Although recent MLE agents have shown promise in automating pipeline design, their behavior has been studied almost exclusively in fully supervised settings. In many domains, however, labeled data is expensive while unlabeled data is abundant. In this work, we study agents in the semi-supervised learning (SSL) regime by framing SSL as a program search problem and exploring how agents navigate this space under limited supervision. Our analysis focuses on three questions: (1) whether agents can effectively leverage unlabeled data, (2) how they leverage it and what strategies they discover, and (3) whether they can adapt these strategies to specific domains. We find that agents can leverage and synthesize SSL strategies spanning decades of SSL research, achieving performance comparable to, and in some cases outperforming, established human-engineered SSL methods. We further validate these observations on real-world ecological datasets, highlighting both the challenges and opportunities of transferring agent-discovered strategies beyond controlled benchmarks.

GPU kernel optimization is increasingly critical for efficient deep learning systems, but writing high-performance kernels still requires substantial low-level expertise. Recent AI coding agents can iteratively read code, invoke compilers and profilers, and refine implementations, yet existing kernel benchmarks evaluate single LLM calls rather than full agent workflows, and none include both kernel-to-kernel optimization and unseen-configuration generalization testing. We present AgentKernelArena, an open-source benchmark for measuring AI coding agents on GPU kernel optimization. The benchmark contains 196 tasks spanning HIP-to-HIP optimization, Triton-to-Triton optimization, and PyTorch-to-HIP translation, and evaluates complete agent workflows in isolated workspaces using gated compilation, correctness, and performance checks, centralized scoring and an unseen-configuration generalization protocol that tests whether optimizations transfer to input configurations the agent never observed. Across production agents including Cursor Agent, Claude Code, and Codex Agent, we find near-perfect compilation and high correctness rates on most task categories, with the strongest configurations achieving mean speedups of up to 6.89$\times$ on PyTorch-to-HIP, 6.69$\times$ on HIP-to-HIP, and 2.13$\times$ on Triton-to-Triton tasks. Our unseen-configuration evaluation shows that HIP-to-HIP and Triton-to-Triton optimizations largely transfer to unseen input shapes, while PyTorch-to-HIP exhibits substantial correctness drops, indicating that agents generating kernels from scratch frequently hardcode shape-specific assumptions. AgentKernelArena is designed as a modular, extensible framework for rigorous evaluation of agentic GPU kernel optimization across agents, tasks, and hardware targets. Code and data are available at \url{https://anonymous.4open.science/r/AgentKernelArena_Neurips2026-F646}.

Probabilistic linear solvers (PLSs) return probability distributions that quantify uncertainty due to limited computation in the solution of linear systems. The literature has traditionally distinguished between Bayesian PLSs, which condition a prior on information obtained from projections of the linear system, and probabilistic iterative methods (PIMs), which lift classical iterative solvers to probability space. In this work we show this dichotomy to be false: Bayesian PLSs are a special case of non-stationary affine PIMs. In addition, we prove that any realistic affine PIM is calibrated. These results motivate a focus on (non-stationary) affine PIMs, but their practical adoption has been limited by the significant manual effort required to implement them. To address this, we introduce *affine tracing*, an algorithmic framework that automatically constructs a PIM from a standard implementation of an affine iterative method by passing symbolic tracers through the computation to build an affine computational graph. We show how this graph can be transformed to compute posterior covariances, and how equality saturation can be used to perform algebraic simplifications required for computation under specific prior choices. We demonstrate the framework by automatically generating a probabilistic multigrid solver and evaluate its performance in the context of Gaussian process approximation.

Temporally extended exploration via graph Laplacian-based options is a promising approach to sparse-reward reinforcement learning (RL), but existing methods either do not explicitly target novelty or fail to scale to pixel-based domains under function approximation. Novel Exploration via Orthogonality (NEO) addresses the first issue by constructing options that navigate from highly visited regions toward less visited ones, yet prior results were limited to settings where exact eigenvectors can be computed. We present a scalable extension of NEO to pixel-based domains, built on three contributions. First, we use a novelty-weighted continuous Laplacian graph-drawing objective, which enables RL with continuous observations. Second, we embed the resulting eigen-potential options within a hierarchical reinforcement learning framework, enabling coherent temporally extended behavior. Third, we observe that learned eigen-potential rewards are directional but locally unreliable under online approximation; we therefore augment each option reward with a novelty bonus, a novel design idea that proves essential for stabilizing option learning while preserving novelty-directed exploration. Together, these contributions yield stronger and more persistent exploration, enabling longer option rollouts and better access to hard-to-reach novel states. Empirically, our method significantly outperforms both the prior scalable Laplacian-option baseline and a direct extension of NEO on sparse-reward benchmarks under a fixed budget. On Montezuma's Revenge, our best variant achieves approximately 1.8x higher return than both baselines. On Venture, both baselines yield returns near zero, whereas our method achieves a return of 1135. Across seven hard ProcGen games, our method achieves approximately 3.5x and 5.6x higher aggregate normalized return than the two baselines, respectively.

The target-charging technique (TCT), a generalization of the sparse vector technique, monitors whether query answers deviate beyond acceptable bounds while incurring privacy costs only upon significant deviations. We develop a more mature theory of TCT through three contributions. First, we generalize TCT and replace prior simulation-based analyses with a direct stochastic domination argument, yielding a tighter bound and a modular proof structure that affords flexibility in the choice of privacy representation for the base mechanism as well as the composition accountant. Second, we provide an anytime-valid TCT guarantee that integrates with approximate Rényi DP filters. Third, we enable individual-level privacy filters. Together, these provide the first end-to-end TCT construction compatible with privacy filters and individual-level accounting, with practical relevance demonstrated through benchmarks inspired by the impending world-scale deployment of the W3C Attribution standard, which is fundamentally rooted in individual filters.

Adaptive Random Forests (ARF) are among the most effective ensemble methods for learning from non-stationary data streams, yet their success relies on heuristic drift detectors and reset rules that lack formal guarantees and require careful tuning. We revisit ARF from an online learning perspective and show that its design decomposes into two fundamental questions: how to aggregate a dynamically evolving set of trees, and how to update the pool of trees over time. We address aggregation using parameter-free online learning methods with strongly adaptive regret, enabling the ensemble to track the best tree mixture. For pool updates, we replace detector-driven resets with deterministic multi-scale scheduling based on geometric lifetimes, combined with incumbent--challenger replacement governed by anytime-valid statistical tests. This design yields theoretical guarantees: global control of false replacements, bounds on promotion delay under post-change advantage, and adaptivity under piece-wise stationary environment. Empirically, the resulting Online RF consistently outperforms ARF and is competitive with broader streaming baselines across several classification and regression benchmarks.

Most active learning (AL) methods estimate sample value through empirical criteria such as prediction ambiguity, diversity, or distributional coverage. We revisit AL from a spectral-geometric perspective by regarding the feature span of the current labeled set as the modeled subspace and its orthogonal complement as an operational residual subspace. The projected residual component of an unlabeled sample then reflects its relation to representation components weakly expressed by the labeled-feature spectrum. Based on this formulation, we propose a spectrally driven nullspace regulation framework that unifies spectrum-induced subspace decomposition, cross-cycle residual persistence estimation, and coverage-aware acquisition in the residual-coordinate space. This yields an acquisition criterion that favors samples whose residual components are both persistent across cycles and complementary to already selected samples. Experiments on visual benchmarks show competitive or improved sample efficiency over strong baselines. Further analyses support the proposed view by showing that residual components shrink anisotropically and that acquisition rankings vary substantially across backbones, indicating that sample value is conditioned on the current representation state rather than being purely data-intrinsic.

In adversarial multi-armed bandits, two performance measures are commonly used: static regret, which compares the learner to the best fixed arm, and dynamic regret, which compares it to the best sequence of arms. While optimal algorithms are known for each measure individually, there is no known algorithm achieving optimal bounds for both simultaneously. \cite{marinov2021pareto} first showed that such simultaneous optimality is impossible against an \emph{adaptive} adversary. Our work takes a first step to demonstrate its possibility against an \emph{oblivious} adversary when losses are \emph{deterministic}. First, we extend the impossibility result of \cite{marinov2021pareto} to the case of deterministic losses. Then, we present an algorithm achieving optimal static and dynamic regret simultaneously against an oblivious adversary. Together, they reveal a fundamental separation between adaptive and oblivious adversaries when multiple regret benchmarks are considered simultaneously. It also provides new insight into the long open problem of simultaneously achieving optimal regret against switching benchmarks of different numbers of switches. Our algorithm uses negative static regret to compensate for the exploration overhead incurred when controlling dynamic regret, and leverages Blackwell approachability to jointly control both regrets. This yields a new model selection procedure for bandits that may be of independent interest.

The internal activations of modern machine learning models align surprisingly well with the activities recorded in real brains. In the last five years, the alignment between encoding models and brain activities, often called brain scores, has steadily increased. Because brain recordings are noisy, there exists an upper bound to the brain scores that any model can achieve. This bound, called the noise ceiling, is crucial for two reasons: it determines whether further improvements in model performance are possible, and it enables meaningful comparisons of brain scores across benchmarks. The field has developed multiple approaches to estimate noise ceilings, but the statistical properties of existing estimators remain poorly understood, leaving it unclear in which experimental settings they provide reliable estimates. Here, we introduce a unified mathematical framework for brain recordings, use it to derive closed-form expressions for four commonly used noise ceiling estimators, and provide theoretical predictions for the behavior of each estimator. We validate the theoretical predictions with simulations that we compare to noise ceiling estimates in existing fMRI datasets. We find that the existing methods underestimate the true noise ceiling in many settings. We conclude with recommendations for best practice in different situations.

Self-distillation enables language models to learn in an on-policy fashion from their own trajectories by using the same model as both the teacher and student, with the teacher being conditioned on privileged information that is unavailable to the student. This information can take multiple forms or views, including solutions, demonstrations, or feedback; for example, for a math problem, one view might be a ground truth reasoning trace. Supervising the student via the privileged teacher enables fine-grained token-level feedback and removes misalignment caused by distillation from external policies with different semantic biases. However, self-distillation also creates a fundamental asymmetry: the teacher may encode view-specific information tied to the privileged information provided during training, which the student cannot access at inference time. Moreover, the best form of privileged information is often task-dependent, making it difficult to choose a single teacher view. In this work, we address both these challenges jointly by introducing AVSD (Adaptive-View Self-Distillation), a novel method of self-distillation with multiple privileged-information views, which reconstructs token-level supervision by separating stable cross-view consensus from view-specific residual support. The consensus signal provides a robust update direction, while the residual signal is selectively used to adjust the update magnitude. Together, these components form a gated aggregation mechanism that balances a conservative target (geometric mean) with a more permissive target (arithmetic mean) for aggregating teacher signals at the token level. Experiments on math competition benchmarks (AIME24, AIME25, and HMMT25) show that AVSD consistently outperforms single-view self-distillation baselines and GRPO, achieving 3.1\% Avg@8 gain over the strongest baseline on Qwen3-8B on average. Moreover, on code-generation benchmarks (Codeforces, LiveCodeBench) using Qwen3-8B, AVSD outperforms single-view self-distillation baselines by 2.9\% on average. More broadly, our analysis shows that AVSD provides better token-level learning signals than any single-view method by reconstructing supervision from multiple privileged teachers.

Audio-video (AV) editing aims to modify audio and video content according to a target prompt. Unlike single-modality editing, AV editing requires models to infer a modality-selective edit scope from the prompt alone: determining not only what should change, but also which modality should be preserved. Faithfully evaluating such models therefore requires both (i) benchmarks that span diverse edit types and modality categories, and (ii) evaluation that is itself modality-aware and sample-specific. However, existing AV editing benchmarks provide limited coverage of edit types and modality combinations, while current evaluation systems are often modality-blind and sample-agnostic, making it difficult to assess whether models faithfully preserve the unintended modality. To address these gaps, we introduce AVENUE, Audio-Video EditiNg Understanding and Evaluation, comprising two contributions: (1) a benchmark of 1,291 source clips and 7,957 editing instructions across audio-targeted, video-targeted, and AV-coupled edit types, curated and human-verified from VGGSound; and (2) a sample-specific, modality-aware evaluation framework that specifies, for each sample, both the intended change and the content that must remain intact. We evaluate representative AV editing models spanning three editing paradigms — joint, sequential, and separate — providing the first systematic analysis of modality-selectivity across paradigms. Our findings reveal a fundamental open challenge: when editing one modality, existing models frequently induce unintended changes in the other, regardless of paradigm. AVENUE provides a benchmark and modality-aware evaluation framework to drive progress toward more controllable AV editing models.

Predictive models have the potential to transform drug development by increasing success rates across all stages of development pipelines. Evaluating this potential needs to include the whole pipeline, from initial screens of thousands of compounds to the expensive assays, such as toxicity assessment, focused on a few compounds in the later stages. Genetic medicines, such as antisense oligonucleotides (ASOs), could benefit millions affected by individually rare genetic disorders, if economic limitations imposed by the need to screen hundreds of sequences at considerable expense can be overcome. To enable systematic analysis of this problem, we present ASO Atlas 2.0, the largest public multi-endpoint ASO preclinical dataset, spanning 295,007 data points, including both in vitro efficacy and in vivo toxicology, across 430. Using this resource, we introduce a cost-based evaluation framework that translates per-stage classifier enrichment into projected cost savings under a representative preclinical pipeline model. We find that incorporating model predictions across all stages reduces projected pipeline cost by 71%. This saving is largely attributable to modest enrichment at costly late-stage in vivo studies rather than strong enrichment at lower-cost in vitro assays, reversing the model ranking suggested by accuracy alone and highlighting the need for improved toxicology modelling.