We study online statistical inference for functionals of the return distribution under a fixed policy. The return distribution is estimated by nonparametric distributional temporal-difference learning from a single Markov trajectory. For the Polyak--Ruppert averaged estimator, we prove that its root-T error converges weakly to a centered Gaussian random element in Cramér space. We also prove that, conditionally on the observed trajectory, the root-T difference between the bootstrap and original averages converges weakly to the same Gaussian limit. These results justify bootstrap inference for smooth statistical functionals, including variance, CVaR, expected shortfall, and expectiles. For nonsmooth statistical functionals, we develop a local asymptotic theory for the estimated return CDF over T^{-1/2}-neighborhoods of finitely many thresholds, together with its bootstrap analogue. This theory allows us to conduct inference for nonsmooth statistical functionals characterized by CDF equations, including return quantiles.
We present a survey of the past and future of AI Scientists: machines capable of automating science. AI Scientists can originate hypotheses, deduce their consequences, design and execute experiments, interpret their results, and revise their beliefs. Such systems are integrated scientific agents, connected to the literature, formal knowledge, mathematical models, simulations, data-analysis systems and physical laboratories. Adam was the first machine to make novel scientific discoveries through cycles of hypothesis formation and physical experimentation. Eve established the architecture of the modern self-driving laboratory. Foundation models, autonomous agents and laboratory robotics now make it possible to build systems far more general than either Adam or Eve. The central problem is no longer whether individual components of science can be automated. They can. The problem is integration. AI Scientists must combine neural learning with logic, probability, mathematics, causal reasoning, simulation, experimental design, robotics and formal scientific records. AI Scientists have the potential to transform science: to make science faster, cheaper, more systematic and more reproducible. AI Scientists could investigate systems too complicated for unaided human science, and enable thousands of AI scientists to work together on single problems. The Nobel Turing Challenge sets the goal of developing by 2050 AI systems capable of automating Nobel-quality discoveries. Progress is ahead of schedule. When we succeed it will create a new form of science and transform the world.
Art style is a signature of professional digital artists that develops through repeated experimentation, reflection, and adaptation. While generative AI (GenAI) can reproduce styles with high fidelity, current tools provide limited support for exploring new stylistic directions and may encourage style replication over exploration. To address this gap, we propose Analyze-Experiment-Resituate (AER), a framework for AI-assisted style exploration derived from interviews with 10 professional digital artists. Rather than prioritizing visually appealing outputs alone, AER supports three core practices of style exploration, including interpreting references, trying out stylistic possibilities, and reflecting on how emerging styles may be received. Specifically, AER enabled artists to (1) analyze artworks into interpretable stylistic elements, (2) have controllable experimentation guided by their own choices, and (3) resituate emerging styles through simulated social perspectives. We implemented AER in a prototype system and evaluated it in a controlled study with 16 artists. Compared with a direct style-transfer workflow, AER increased artists' agency and reflection as they pursued new stylistic directions. A two-week field study with four artists revealed how the AER framework influenced daily style exploration, such as reflection, experimentation, and stylistic decision-making at each stage. We discuss opportunities and challenges in designing AI-assisted style-exploration workflows, and outline implications for future artistic support tools.
In offline RL, estimating the optimal action-value function Q^* can be formulated as solving the optimal Bellman equation based solely on offline observations. A fundamental challenge is that the reward function and transition kernel are unknown, so the optimal Bellman operator is not directly observable from data. To address this issue, we propose a novel framework that decouples operator estimation from value function learning. In this approach, we first formulate conditional diffusion models to estimate the reward law and transition kernel, which induces a data-driven approximation of the optimal Bellman operator. We then plug these estimators into the Bellman equation and obtain a deep estimator of Q^* by minimizing the empirical Bellman residual over a neural network function class. Theoretically, we first establish sharp nonasymptotic convergence rates for learning the optimal Bellman operator through an end-to-end analysis of conditional diffusion estimation in total variation distance. We then establish the oracle value-stage rate {O}(n^{-{2β}{d_x+d_a+2β}}) for the excess Bellman residual risk. Finally, under a concentrability condition, we translate this residual bound into an L^2 convergence rate of {O}(n^{-β{d_x+d_a+2β}}) for the resulting deep estimator of Q^*, where d_x and d_a denote the dimensions of the state and action spaces, respectively, and β denotes the Hölder smoothness index of Q^*. Importantly, our theoretical analysis does not rely on completeness assumptions commonly used in deep RL theory. Extensive numerical experiments demonstrate the effectiveness of the proposed method and its strong empirical performance.
Patients increasingly ask large language model (LLM) assistants which doctor to see, making these systems AI infomediaries: algorithms that intermediate one person's choice among other people and thereby decide, silently and at scale, which physicians become visible. We report a prespecified randomized algorithm audit of what causally moves those recommendations. Seven models (six open-weight; gpt-4o-mini) each chose among five synthetic family-medicine physician cards whose attributes were independently randomized across 3,024 choice sets, three patient personas, nine prompt paraphrases and nine experimental arms, yielding 40,068 scored responses; gender and ethnicity were signaled through names following correspondence-audit methodology. Reputation signals dominate: raising a rating from 3.9 to 4.7 increases choice probability by 31.4 percentage points (pp), and raising the fee from 90 to 190 lowers it by 20.0 pp. Demographic parity is rejected, but not in the direction human audit studies predict: female-signaled names gain 2.5 pp, and Hispanic-, South-Asian- and Black-signaled names gain 1.3-2.9 pp over White-signaled names, tilts worth 7-14 per visit in fee-equivalent terms, and a content-free first-listed position is worth $11. Yet models mentioned gender or ethnicity in at most 0.03% of their stated reasons and abstained in 0.39% of trials, so these effects are invisible in the models' own explanations, and transparency obligations relying on model self-report would not detect them. One reasoning model failed the prespecified auditability gate outright. The frozen design makes the audit repeatable: any new model can be assessed against identical stimuli, making recurring behavioural audit, rather than self-reported explanation, the monitoring technology fit for purpose.
Zahavy [2026] argues that Large Language Models, despite their capabilities in induction and deduction, cannot perform the abductive "Jump" that produced Einstein's equivalence principle, and attributes this limitation to the absence of embodied simulation. Zheng-Xin [2026] and Farmer [2026] question whether embodiment is necessary for abduction, pointing to alternative routes to General Relativity and forms of abduction that require no sensorimotor grounding. Max Planck resolved the blackbody radiation problem in 1900. Planck's move to E = hν required no embodied simulation. It was motivated by a mathematical consequence of classical theory, an infinite predicted energy for a finite measured quantity, that could not be physically accepted. We show that neither induction nor deduction could have produced the postulate and argue that its adoption required a coupling between epistemic error and physical cost. We formalize this distinction through thermodynamic coupling and show that fixed-weight transformer inference lacks such coupling, regardless of model scale. This is consistent with empirical results showing that output entropy remains nearly unchanged across tasks with sharply increasing causal difficulty, even as accuracy falls from 100% to 17%. We therefore argue that the missing ingredient in machine abduction may lie deeper than embodiment: a system must have some physical mechanism through which epistemic error becomes costly enough to force revision.
The alternating direction method of multipliers (ADMM), as a landmark algorithm, has attracted tremendous research attention and extensive practical applications over the past two decades. It is well known that, although the two-block ADMM enjoys well-established theoretical convergence guarantees, its direct extension to the three-block case may fail to converge, as demonstrated by existing counterexamples [5]. However, to the best of our knowledge, the case in which the third constraint block is the identity remains unresolved: the existing literature gives neither a general convergence proof nor a counterexample for this subclass. In this paper, we give a negative answer: direct three-block ADMM may fail even when the first two blocks are strongly convex quadratics. Using Codex with GPT-5.6 Sol, we construct an explicit rational counterexample candidate and verify it along a piecewise-affine reduction path; exact checks show that direct three-block ADMM on this instance produces a bounded nonconvergent orbit of period 66. Within the same Codex workflow, we further guide a study of multiplier relaxation and clarify when convergence can be restored at the fixed-instance and class levels: a problem-dependent small dual step can restore convergence, whereas no positive relative step works uniformly over the whole class. Furthermore, we also test the recent Kimi Code with Kimi K3 model without the Codex candidate or project-specific route guidance; along a different path it produces an exact locally attracting period-23 certificate, convertible to an equivalent all-identity instance. The comparison suggests that different research-harness configurations can shape the mathematical objects explored and the certificates pursued.
Neuron- and path-level interventions offer the finest-grained route to defending large language models (LLMs) against jailbreak attacks, yet existing methods fall short of this promise, i.e., they often compromise model utility significantly. Specifically, one line of work suppresses toxic neurons to erase harmful semantics, but since such semantics are distributed across the network, blocking every pathway forces a large intervention footprint. An alternative line of research focus on identify safety neurons using external classifiers. While promising, the existing approaches suffer from compromising neurons that are important for the model utility as well. Moreover, both approaches remain always on and thus perturb every benign request even when no attack is present. To address these limitations, we present {}, a training-free defense that first identifies safety-specific neurons through per-neuron hypothesis tests under false-discovery-rate control together with a utility-specificity filter. Based on this identification, a trigger-style clamp holds the selected neurons at their harmful-conditional mean activations, injecting an internal harmful-input signal that triggers the refusal behavior learned during alignment. The clamp is then realized by two provably equivalent deployment modes, namely a detector-gated inference-time intervention and an offline bias-patch weight edit. Extensive experiments across four safety-aligned LLMs and four representative attacks demonstrate that {} reduces the average attack success rate to at most 2.0\% while incurring a utility drop of only 0.5\% to 5.3\% on MT-Bench, the smallest among all defenses. Code is available at https://anonymous.4open.science/r/Tripwire-65C4.
Recent video generators can fabricate realistic depictions of wars, disasters, public emergencies, and other real-world crises, creating substantial risks of misinformation. Existing benchmarks, however, provide limited evidence on detector and generator behavior in such settings, including how detectability varies with generation conditions, how people perceive generated videos, and whether detectors remain reliable during social dissemination. To address this gap, we introduce RA-Bench, a benchmark for AI-generated video detection that uses Real videos as Anchors. RA-Bench contains 17,886 videos, comprising 1,830 real-video anchors across 10 social-risk categories and 16,056 generated clips from four open-source and five closed-source generators. Based on RA-Bench, we organize our evaluation along three dimensions. We first assess detector generalization across seven traditional detectors, ten zero-shot multimodal models under three review settings, and two MLLMs specifically fine-tuned on AI-generated video detection. Across these methods, none of the three detector families generalizes consistently across RA-Bench instances. We then examine how detectability varies with generation quality, conditioning information, and sampling seeds. These analyses show that generation properties affect detector families differently, while source-level detection patterns remain stable across seeds. Finally, we study human authenticity judgments and detector reliability during social dissemination. We find that videos that mislead people are also difficult for current detectors, and that social dissemination makes detection harder. Together, these findings show that current methods struggle to detect realistic AI-generated videos, highlighting the need for detectors robust to evolving video generators.
Existing palm presentation attack detection (PAD) datasets are often limited by static imagery, restricted acquisition conditions, or insufficient multimodal video data, hindering systematic evaluation across environments, modalities, and attack types. We present GBU-Palm, a large-scale multimodal video dataset and benchmark containing 21,326 videos from 105 subjects and 210 palms across six acquisition environments, including bona fide, Print, and Replay presentations, with 6,310 synchronized RGB-NIR samples. We construct leakage-controlled protocols that separate palm identity and attack lineage and benchmark four representative video architectures under environment-matched and held-out-environment settings. Results reveal substantial architecture-dependent degradation under environmental shift and show that RGB-NIR fusion does not consistently outperform RGB-only input. We further analyze model behavior through true accept (TA), true reject (TR), false accept (FA), and false reject (FR) decomposition, spectral masking, temporal-order intervention, and frozen-backbone NIR probing, revealing distinct failure patterns and evidence utilization across architectures. GBU-Palm provides a unified and challenging benchmark for developing and evaluating robust multimodal palm PAD methods under cross-environment conditions.
Mixture-of-Experts (MoE) models have been widely adopted in real-time interactive applications such as coding assistants, real-time audio-video interaction systems. To meet the extremely low response latency requirements of these scenarios, practitioners commonly employ small-batch decoding, under which MoE inference becomes memory-bound and is severely bottlenecked by expert weight loading. However, this bottleneck has received limited attention, and existing solutions such as post-training weight compression or fine-grained expert design during pre-training either degrade model accuracy or introduce additional computation and communication overhead. To tackle this issue, we propose DeaMoE, a decoding-efficient MoE architecture, in which the experts are grouped into several departments, and the experts belonging to the same department share most parameters since they come from the same professional field, and additionally each expert contains a few private parameters to reflect its uniqueness. Moreover, we design customized two-stage routing strategy for DeaMoE to avoid redundant loading, under which DeaMoE greatly improves the efficiency during LLM decoding. Compared with vanilla MoE, DeaMoE reduces per-step loaded weights by up to 50.9% and achieves up to 1.33 end-to-end TPOT speedup for the pre-trained 7B model on A40, and up to 2.00x and 1.97x peak speedup for DeepSeek-V3 on A40 and H100 in microbenchmarks.
Many real-world tasks require LLM agents to interact with their environments over long execution horizons. Errors that occur early in execution may propagate through both the agent context and environment state, and their effects may be difficult to reverse through subsequent actions. Existing methods mainly seek to reduce such errors through plan refinement and safety checks but provide little support after errors occur. To enable recovery during long-horizon execution, we present AgentRewind, a runtime recovery framework that records aligned checkpoints of the agent context and controlled environment, allowing agents to return to an earlier state and resume execution with information from previous attempts. We also construct MettleBench, a benchmark for evaluating task completion and partial progress on long-horizon engineering assignments containing a series of related requirements. Experiments across tasks, multiple models, execution strategies, and agent harnesses show that AgentRewind improves task success rate and average checklist progress over the compared baselines.
Vision-Language-Action (VLA) models have recently achieved promising performance in robotic manipulation. However, existing benchmarks mainly evaluate generalization on static manipulation tasks and largely overlook dynamic interaction scenarios. To address this gap, we present ReflexBench, a benchmark for reaction-critical manipulation. ReflexBench contains six dynamic tasks and introduces an evaluation framework that decouples simulator stepping from robot control while supporting configurable latency under synchronous and asynchronous inference. Building upon ReflexBench, we propose ReflexVLA, an efficient VLA model designed for reaction-critical manipulation without large-scale robot-data pretraining. ReflexVLA enhances temporal reasoning through latent future prediction and multi-frame temporal fusion within the vision backbone, while reducing deployment latency through batched visual encoding and CUDA Graph replay. Experiments show that ReflexVLA consistently improves dynamic manipulation performance while maintaining competitive accuracy on standard static manipulation benchmarks, and real-world experiments further demonstrate its effectiveness under practical deployment conditions. Project website: https://reflexvla.github.io
Sports have witnessed growing global enthusiasm in recent years, serving as a vital force for physical health, cultural exchange, social connection, and economic growth. The rapid advancement of large models, particularly (multimodal) large language models (M)LLMs, has demonstrated transformative potential to reshape sports understanding, analysis, and interaction across diverse domains. This paper presents a comprehensive survey of large models in sports, including (i) an overview of tasks and applications across different participant groups; (ii) a detailed analysis of sports-related datasets and benchmarks; and (iii) a critical discussion of current challenges and future directions. Our goal is to establish a foundation for advancing research and practical development of large-model-driven sports intelligence. An open-source GitHub repository is maintained at: https://github.com/Road2Redemption/Awesome_Large_Models_In_Sports1.
Multi-agent reasoning systems often use agreement, confidence, or automated scores to decide which messages should shape a final answer. Such filtering assumes that a message likely to be correct is also worth keeping. Yet a wrong answer can contain a useful decomposition, constraint, or scientific principle. We test this distinction with Diverse Hypothesis Deliberation (DHD), a controlled measurement protocol that caches five independently generated messages and replays the same downstream solver, called the integrator, with each message available or hidden. The replay comparison measures a message's trajectory value: whether making the message available helps or harms subsequent reasoning. Across five mathematics and science benchmarks and two openly available model families, gpt-oss-120b and gemma-4-31B-it, wrong-helpful messages appear in every benchmark-model combination. Among wrong-answer messages that change final correctness, more than four in ten changes are helpful in each model. Controlled repeats show that the number of repeatable message effects is unlikely to arise from replay variation alone (p=0.0002). A focused intervention on repeatable wrong-helpful messages finds that the complete message works best, while retaining its reasoning preserves more success than retaining only its answer; the source of the complete-message advantage remains open. Within the same problem, repeated trajectory-value evidence also identifies a better keep-or-remove choice than answer correctness alone. Answer correctness is therefore informative but does not determine trajectory value. DHD measures this missing property and produces reusable labels for learning when agents should listen.
The symmetries of a learning task have become an important factor in designing modern deep learning solutions. Data augmentation is a straightforward and effective way of incorporating symmetries into a generic neural network. Recent results show that infinitely large deep ensembles show perfect symmetry when trained on augmented data. However, since training ensembles requires repeating the training process many times, this method is costly. In this work, we study stochastic weight averaging (SWA) as an alternative ensembling technique that does not require repeated training runs. We analyze SWA by approximating the stochastic training trajectory at the end of training with an Ornstein--Uhlenbeck process. We show that in the infinite-width limit, SWA on augmented data provides an equiviariance boost that goes beyond what could be expected from the performance increase due to SWA alone. We verify our results with extensive numerical experiments on numerous models spanning computer vision and graph classification with both discrete and continuous symmetries.
Scientific data often describe entities whose features are jointly governed by the laws of physics, yet existing self-supervised learning (SSL) objectives largely ignore this physical coherence. We introduce imposter, a discriminative pretext task that replaces subsets of an entity's features with real observations donated by another entity and trains the encoder to identify the swapped features. Because every donated value is individually plausible, the task can only be solved by learning cross-feature physical dependencies. We evaluate the proposed objectives on global ERA5-Land reanalysis data using 21 environmental variables and assess the learned representations on seven downstream tasks spanning climate classification, carbon flux estimation, and streamflow prediction. Our study includes, to our knowledge, the first systematic comparison of self-supervised objectives for land-surface modeling under a shared architecture and pre-training budget. We find that the most effective pretext task depends on the downstream task family rather than any single objective's superiority, and that imposter provides complementary information when combined with existing SSL objectives. These results suggest that physical coherence is a valuable new source of self-supervision for scientific foundation models.
The modelling and analysis of secure business processes require the incorporation of security annotations into process models. Although BPMN extensions, including SecBPMN2, exist for this purpose, the derivation of accurate and complete security annotations from natural-language specifications remains a manual, expert-intensive, and error-prone task. This paper presents a hybrid framework that takes a BPMN process model and a security requirements document as input and automatically generates security annotations adhering to the SecBPMN2 specification. The approach combines Large Language Model (LLM)--based semantic extraction with schema-constrained mapping, rule-based normalization, and deterministic validation. The framework is evaluated comprehensively on a curated dataset of 27 process models from various domains. The results indicate that it consistently produces structurally valid SecBPMN2 annotations with high schema completeness. Compared to human security analysts, the system achieves substantially higher precision (0.58 vs. 0.29) while maintaining comparable recall (0.52 vs. 0.50) and reduces erroneous or misplaced annotations by nearly 50%. In addition, annotation generation is significantly faster than manual annotation. These findings demonstrate that hybrid LLM- and rule-based automation can reduce modeling effort while improving consistency and reliability, thereby providing a scalable foundation for security-by-design BPM.
