Text-to-motion (T2M) generation has broad applications in character animation, virtual avatars, and human-robot interaction. Existing methods typically generate pose trajectories or motion tokens directly from language, forcing a single model to handle semantic interpretation, long-horizon structure, and low-level physical realization. This coupling makes them costly and often unreliable for long, compositional, or semantically dense prompts. We propose Text2BFM, the first framework that aligns natural language with pretrained Behavioral Foundation Models (BFMs) for T2M generation without relying on heavy end-to-end motion generators. Text2BFM operates in the latent policy space of a frozen BFM, using it as an executable motion prior. A text-aligned variational behavioral bottleneck compresses BFM policy-latent sequences into compact motion representations that are compatible with language and preserve long-horizon behavioral structure. Generation is performed in this compact behavioral manifold with a lightweight conditional generator, and the resulting latent encoded behaviors are decoded into policy latents that drive the pretrained frozen BFM. By decoupling semantic planning from motion execution, Text2BFM achieves efficient, robust T2M generation and strong performance on long, compositional textual descriptions.

Data-free knowledge distillation-based one-shot federated learning (OSFL) trains a model in a single communication round without sharing raw data, making OSFL attractive for privacy-sensitive medical applications. However, existing methods aggregate predictions from all clients to form a global teacher. Under non-IID data, conflicting predictions dilute each other during averaging, yielding less informative soft labels that weaken distillation. We propose FedBiCross, a personalized OSFL framework with three stages: (1) clustering clients by model output similarity to form coherent sub-ensembles, (2) bi-level cross-cluster optimization that learns adaptive weights to selectively leverage beneficial cross-cluster knowledge while suppressing negative transfer, and (3) personalized distillation for client-specific adaptation. Experiments on four medical image datasets demonstrate that FedBiCross consistently outperforms state-of-the-art baselines across different non-IID degrees.

Proteins inherently possess a consistent sequence-structure duality. The abundance of protein sequence data, which can be readily represented as discrete tokens, has driven fruitful developments in protein language models (pLMs). A key remaining challenge, however, is how to effectively integrate continuous structural knowledge into pLMs. Current methods often discretize protein structures to accommodate the language modeling framework, which inevitably results in the loss of fine-grained information and limits the performance potential of multimodal pLMs. In this paper, we argue that such concerns can be circumvented: a sequence-based pLM can be extended to incorporate the structure modality through continuous tokens, i.e., high-fidelity protein structure latents that avoid vector quantization. Specifically, we propose a hybrid diffusion protein language model, HD-Prot, which embeds a continuous-valued diffusion head atop a discrete pLM, enabling seamless operation with both discrete and continuous tokens for joint sequence-structure modeling. It captures inter-token dependencies across modalities through a unified absorbing diffusion process, and estimates per-token distributions via categorical prediction for sequences and continuous diffusion for structures. Extensive results demonstrate that HD-Prot achieves competitive performance in unconditional sequence-structure co-generation, motif-scaffolding, protein structure prediction, and inverse folding tasks. Furthermore, our method can perform on par with state-of-the-art multimodal pLMs, despite being developed under limited computational resources (i.e., less than one-tenth the budget for modality extension fine-tuning). It highlights the viability of simultaneously estimating categorical and continuous distributions within a unified language model architecture, offering a promising alternative direction for multimodal pLMs.

We prove real-rootedness for the Poincaré polynomial \[ P_n(t)=\sum_{i=0}^{n-3} \dim H^{2i}(\overline{\mathcal M}_{0,n};\mathbb{Q})t^i \] of the Deligne--Mumford moduli space $\overline{\mathcal M}_{0,n}$ of stable $n$-pointed rational curves, proving a conjecture of Aluffi--Chen--Marcolli. The proof starts from the Keel--Manin--Getzler recurrence, but its main new idea is a bivariate deformation $F_m(y,t)$ of the Poincaré polynomial. This deformation reveals a hidden interlacing structure not visible in the one-variable recurrence. For fixed $t<0$, the zero set of $F_m$ in the $y$-direction is controlled by a Sturm--Rolle argument on the interval $0<y<1-t$. The original polynomial is recovered on the slice $y=1$, and the ordered crossings of the moving roots through this slice give both real-rootedness and strict interlacing. Consequently, the Betti numbers of $\overline{\mathcal M}_{0,n}$ form an ultra-log-concave sequence. We further prove real-rootedness and ultra-log-concavity for the Poincaré polynomial of the Fulton--MacPherson space $\mathbb{P}^1[n]$ of $n$ ordered points in degenerations of the complex projective line. The proof for $\overline{\mathcal M}_{0,n}$ was obtained through an iterative AI-assisted workflow with Co-Mathematician, an agentic frontier-model system developed by Google DeepMind. Our role was to formulate the problem, evaluate the proposed proof attempts, identify gaps and request corrections, compare the developing argument with the literature, and refine the presentation of the final proof. Our additional human contribution was to observe that a similar residual deformation strategy applies to the Fulton--MacPherson spaces $\mathbb P^1[n]$, yielding the corresponding real-rootedness theorem.

Mechanistic interpretability often studies Transformer behavior by intervening on internal activations through activation patching, causal tracing, path patching, and steering directions. This paper develops Transformer Field Theory: a response-theoretic framework in which the residual stream of a fixed forward pass is treated as a Transformer field over layer depth and token position. In this formulation, patching becomes a localized source insertion into the Transformer field, first-order sensitivity fields predict patch effects, Green functions describe downstream propagation, and patch selection is posed as an adjoint inverse problem. Empirically, we test the theory's forward response objects in GPT-2-style autoregressive Transformers. Localized Transformer-field interventions exhibit a bounded local linear regime; first-order sensitivities predict patch effects across layer-token sites; localized sources generate structured anisotropic Transformer-field propagation; high-sensitivity sites and sliced Green operators provide reduced response descriptions; and prompt-induced Transformer-field displacements partially transfer answer behavior. These results establish sensitivities, Transformer-field responses, and sliced Green operators as practical objects for organizing patching experiments, while providing the forward mathematical basis for patch-site inference and cross-scale response transfer.

Time series (TS) reasoning models (TSRMs) have shown promising capabilities in general domains, yet they consistently fail in the financial domain, which exhibits unique characteristics. We propose a general 2 x 2 capability taxonomy for TSRMs by crossing 1) single-entity vs. multi-entity analysis with 2) assessment of the current state vs. prediction of future behavior. We instantiate this taxonomy in the financial domain-where the distinction between deterministic assessment and stochastic prediction is particularly critical-as ten financial reasoning tasks, forming the FinTSR-Bench benchmark based on S&P stocks. To this end, we propose FinSTaR (Financial Time Series Thinking and Reasoning), trained on FinTSR-Bench with distinct chain-of-thought (CoT) strategies tailored to each category. For assessment, which is deterministic (i.e., computable from observable data), we employ Compute-in-CoT, a programmatic CoT that enables models to derive answers directly from raw prices. For prediction, which is inherently stochastic (i.e., subject to unobservable factors), we adopt Scenario-Aware CoT, which generates diverse scenarios before making a judgment, mirroring how financial analysts reason under uncertainty. The proposed method achieves 78.9% average accuracy on FinTSR-Bench, substantially outperforming LLM and TSRM baselines. Furthermore, we show that the four capability categories are complementary and mutually reinforcing through joint training, and that Scenario-Aware CoT consistently improves prediction accuracy over standard CoT. Code is available at this https URL.

Content moderation in online multiplayer 3D virtual environments is increasingly automated, yet detection has focused on images, video, and audio, leaving suggestive motion a blind spot. We present a motion-only classification pipeline that detects suggestive and explicit movement from SMPL skeleton trajectories using Laban Movement Analysis (LMA) descriptors. On a dataset spanning everyday, artistic, suggestive, and explicit movement (17+ hours of video), a logistic regression trained on 61-feature LMA descriptors reaches 68% binary SFW/NSFW accuracy (70% random forest) under a leak-free evaluation protocol. At this level, our descriptor performs comparably to a learned video model trained on the same motion re-rendered as appearance-free video, a gray figure with no clothing, skin, or scene. The indirectness (tortuosity) of each joint's trajectory, measured as the ratio of the joint's path length to its net displacement, peaks at the suggestive tier, showing that the Direct-to-Indirect polarity of Laban's Space factor provides an interpretable marker of the shift from functional to suggestive motion. Ultimately, Laban-based kinematic descriptors offer a lightweight, interpretable approach to suggestive-motion detection: every decision decomposes into named, theory-grounded features. Because the classifier operates on pose trajectories alone, moderation can run directly on avatar poses in virtual environments, with no appearance data.

The advancement of Large Reasoning Models (LRMs) has catalyzed a paradigm shift from reactive ``fast thinking'' text generation to systematic, step-by-step ``slow thinking'' reasoning, unlocking state-of-the-art performance in complex mathematical and logical tasks. However, the field faces \textit{the fundamental gap between token-level behavioral analysis and internal reasoning mechanisms, and the instability of reinforcement learning (RL) for reasoning optimization relying on costly external verifiers}. We identify and formally define \textbf{Entropy-Gradient Inversion}, a robust negative correlation between token entropy and logit gradients that acts as a definitive geometric fingerprint for LRM reasoning capability. Building on this, we propose \textbf{Correlation-Regularized Group Policy Optimization (CorR-PO)}, which embeds this inversion signature into RL reward regularization. Extensive experiments on various reasoning benchmarks across multiple model scales show CorR-PO consistently outperforms state-of-the-art baselines, confirming that stronger inversion directly correlates with superior reasoning performance.

Detecting Schwartz values in political text is difficult because implicit cues often depend on surrounding arguments and fine-grained distinctions between neighboring values. We study when context and explicit moral knowledge help sentence-level value detection. Using the ValuesML/Touché ValueEval format, we compare sentence, window, and full-document inputs; no-RAG and retrieval-augmented settings with a curated moral knowledge base; supervised DeBERTa-v3-base/large encoders; and zero-shot LLMs from 12B to 123B parameters. The results show that more context is not uniformly better: full-document context improves supervised DeBERTa encoders by 3.8-4.8 macro-F1 points over sentence-only input, but does not consistently help zero-shot LLMs. Retrieved moral knowledge is more consistently useful in matched comparisons, improving each tested model family and context condition under early fusion. However, scaling from DeBERTa-v3-base to large and from 12B to larger LLMs does not guarantee gains, and simple early fusion outperforms the tested late-fusion and cross-attention RAG variants for encoders. Per-value analyses show that context and retrieval help most for socially situated or conceptually confusable values. These findings suggest that value-sensitive NLP should evaluate context, knowledge, and model family jointly rather than treating longer inputs or larger models as universal improvements.

In recent years, image editing models have made significant progress, enabling users to manipulate visual content in a flexible and interactive manner through natural language instructions. However, an important yet underexplored research direction remains dense visual document image editing, which involves modifying textual content within images while faithfully preserving the original text style and background context. Existing methods primarily focus on English scenarios and images with relatively sparse text, and thus cannot adequately address dense, structurally complex documents or non-Latin scripts such as Chinese. To bridge this gap, we propose VDE Bench (Visual Doc Edit Bench), a rigorously human annotated and evaluated benchmark specifically designed to assess the performance of image editing models on bilingual Chinese-English and complex visual document editing tasks. The benchmark comprises a high quality dataset of 942 instruction based image editing samples, whose seed images encompass dense Chinese and English text documents including academic papers, posters, presentation slides, examination materials, and newspapers. Furthermore, we introduce a novel evaluation framework that systematically quantifies editing performance at the OCR parsing level, thereby enabling fine grained assessment of text modification accuracy. Based on this benchmark, we conduct a comprehensive evaluation of representative image editing models. Human verification demonstrates a high degree of consistency between human judgments and automated evaluation metrics. VDE Bench constitutes the first systematic benchmark for evaluating the performance of image editing models on bilingual dense text visual documents.

Rapid progress in aerodynamic shape optimization (ASO) has outpaced currently-available standardized evaluation frameworks. Fair comparison requires a unified benchmark spanning diverse shape classes, objective formulations, and matched-budget state-of-the-art baselines. We introduce ShapeBench, an open-source ASO benchmark with a unified API spanning 103 tasks across eight shape categories and multiple optimization regimes. Each ShapeBench task includes a validated surrogate for fast search; when feasible, a high-fidelity Computational Fluid Dynamics (CFD) pipeline for final verification is available, enabling systematic fidelity-gap analysis. ShapeBench provides a reproducible protocol with well-configured baselines to compare fairly using a consistent budget metric, allowing for comparison among both classical and LLM-driven methods, including general-purpose optimizers and a new domain-specialized evolutionary LLM baseline, ShapeEvolve. Results on ShapeBench demonstrate substantial variance in optimizer rankings across shape categories and problem formulations, with mean pairwise Spearman $\rho = 0.013$, so single-task conclusions do not reliably generalize across problem classes. The benchmark is also far from saturation; classical methods are rarely applicable across all shape categories and tasks, further highlighting the need for more general-purpose approaches.

Vision-Language Models (VLMs) hallucinate objects that are not present, and a growing line of work tries to curb this by feeding the model its own generated caption as auxiliary evidence -- assuming that a caption, once available, is something to consume. We show this fails: naively appending a caption can lower accuracy rather than raise it, dropping Qwen2.5-VL-3B$^\dagger$ on HallusionBench by nearly ten points. To understand why, we build \textbf{GD-Probe}, a diagnostic set that pairs a global and a detail question on the same image, so that any difference in caption effect is attributable to the question alone. Caption utility proves to be a \emph{per-query} property: the same caption helps global questions and harms detail ones, through a single mechanism -- an embedded caption competes with the image for attention and pulls the model's evidence onto its own text -- whose sign is set by whether the caption \emph{covers} the queried content. Crucially, this regime is readable from quantities the decoder already emits, with no attention access or grounding. We turn this into \textbf{GEASS} (Gated Evidence-Adaptive Selective Caption Trust), a training-free, logit-level module that decides per query how much of the caption to trust, gating it by the clean path's confidence, weighting it by the entropy reduction it induces, and raising the evidence bar when the two pathways disagree. Across four VLMs and two benchmarks (POPE and HallusionBench), GEASS improves over both vanilla inference and contrastive decoding under a single fixed setting, adding only two forward passes and no parameters.

Diffusion Language Models (DLMs) generate text by iteratively denoising masked token sequences, offering a tradeoff between parallelism and quality compared to autoregressive models. In current practice, the number of tokens decoded per step is controlled by a confidence threshold, and quality degrades monotonically as more tokens are denoised per step. We introduce Multi-token Residual Prediction (MRP), a lightweight module that enables dependency-aware multi-token denoising within a single backbone forward pass. MRP exploits a key property of the denoising process: the logit distributions at adjacent denoising steps are remarkably similar. Rather than running the backbone a second time to obtain the next-step logits, MRP predicts the residual between steps from the backbone's hidden states, effectively denoising more tokens per backbone forward at a fraction of the cost. We apply MRP across the two operating regimes of DLM decoding. In the high-quality-low-throughput static denoising regime, MRP serves as a drafter for speculative decoding: its proposals are verified against the backbone, yielding lossless acceleration of up to 1.4x in SGLang. In the low-quality-high-throughput dynamic denoising regime, MRP instead drives a remasking scheme that revokes over-eager reveals, recovering most of the accuracy lost to aggressive low-threshold decoding and improving accuracy by up to 22.6 points on code generation task HumanEval and 17.7 points on reasoning task GSM8K.

Binary function classifiers play a crucial role in maintaining the security and integrity of software systems by detecting malicious code and unauthorized modifications. However, machine learning-based classifiers are vulnerable to adversarial attacks that can evade detection. In this study, we present Kelpie, a novel framework for executing mimicry attacks, a stronger type of targeted evasion attacks, on binary function classifiers in a black-box, zero-query setting. Unlike previous approaches that rely on querying the target classifier to refine untargeted evasion attacks, Kelpie leverages code transformations that preserve the functionality of malicious payloads while causing them to be misclassified as we want. Through extensive experimentation, we demonstrate that Kelpie can successfully execute mimicry attacks against six state-of-the-art binary function classifiers representing different model architectures without requiring direct interaction with them. We further validate our approach with a practical demonstration, involving a keylogger and a wiper concealed within benign-looking functions embedded in an application. This work, to our best knowledge, is the first to demonstrate such a mimicry attack in a black-box, zero-query context, raising important questions about the reliability and security of existing machine learning-based binary function classifiers.

Continual reinforcement learning challenges agents to acquire new skills while retaining previously learned ones with the goal of improving performance in both past and future tasks. Most existing approaches rely on model-free methods with replay buffers to mitigate catastrophic forgetting; however, these solutions often face significant scalability challenges due to large memory demands. Drawing inspiration from neuroscience, where the brain replays experiences to a predictive World Model rather than directly to the policy, we present ARROW (Augmented Replay for RObust World models), a model-based continual RL algorithm that extends DreamerV3 with a memory-efficient, distribution-matching replay buffer. Unlike standard fixed-size FIFO buffers, ARROW maintains two complementary buffers: a short-term buffer for recent experiences and a long-term buffer that preserves task diversity through intelligent sampling. We evaluate ARROW on two challenging continual RL settings: Tasks without shared structure (Atari), and tasks with shared structure, where knowledge transfer is possible (Procgen CoinRun variants). Compared to model-free and model-based baselines with replay buffers of the same-size, ARROW demonstrates substantially less forgetting on tasks without shared structure, while maintaining comparable forward transfer. Our findings highlight the potential of model-based RL and bio-inspired approaches for continual reinforcement learning, warranting further research.

Spracklen et al. (USENIX Security '25) showed that code-generating large language models hallucinate package names that do not exist on PyPI or npm at rates ranging from 5.2% on commercial models to 21.7% on open-source models, creating an attack surface for slopsquatting -- the registration of malicious packages under hallucinated names. We replicate their methodology on five frontier code-capable LLMs released between October 2025 and March 2026: Claude Sonnet 4.6, Claude Haiku 4.5, GPT-5.4-mini, Gemini 2.5 Pro, and DeepSeek V3.2. Across 199,845 paired Python and JavaScript prompts validated against PyPI and npm master lists, we measure overall hallucination rates between 4.62% (Claude Haiku 4.5) and 6.10% (GPT-5.4-mini) -- an order-of-magnitude compression of the inter-model spread observed by Spracklen, but not a retirement of the threat. Beyond replication, we identify a set of 127 package names (109 on PyPI, 18 on npm) that all five evaluated models invent identically; following coordinated disclosure with PyPI Security and this http URL, 53 of these (41 on PyPI, 12 on npm) remain registrable by an attacker after each registry's existing defenses, constituting a model-agnostic supply-chain attack surface that no single-model study can reveal. We further document a Python-over-JavaScript hallucination asymmetry that inverts Spracklen's 2024 finding, identify a Haiku-below-Sonnet inversion within the Anthropic family, and observe a Jaccard-similarity peak between DeepSeek V3.2 and GPT-5.4-mini (J = 0.343) suggestive of shared training-data origins.

Modern Vision-Language Models (VLMs) achieve strong semantic recognition, yet remain brittle on elementary spatial relations such as left of, on, behind, and between. One cause of this failure arises before language reasoning begins: the visual pathway may compress or discard critical 3D structural cues during feature extraction, so the language model receives image representations that are already insufficient for reliable spatial judgment. We introduce GeoWorld-VLM, a VLM-side distillation framework that transfers geometric structure from frozen camera-conditioned video world models into VLMs. GeoWorld-VLM fine-tunes only the image encoder and multimodal projector, aligning post-projector image features with intermediate world-model representations while leaving the main backbone frozen. Given images, a prompt, and a sampled camera trajectory, the world-model teacher converts static visual input into a synthetic multi-view spatial signal. Training combines spatial answer supervision, teacher-student feature alignment, and a preservation anchor to the original VLM. Since the language model remains frozen, GeoWorld-VLM preserves the original model's linguistic capabilities while attributing spatial improvements to the enhanced visual pathway. To evaluate the effectiveness and generality of the proposed method, we apply GeoWorld-VLM to two distinct VLM architectures and observe consistent improvements across both backbones. GeoWorld-VLM improves performance by approximately 4 percent on both the What'sUp and VSR benchmarks, suggesting that world-model-guided visual alignment generalizes across model structures and spatial reasoning datasets.

Scaling LLMs requires tremendous computational resources, and recent advances in AI have gone hand in hand with massive amounts of capital expenditure. While it is established that scaling up LLMs reliably increases model quality (quantified in terms of loss or downstream evaluations), it is unclear how these quality improvements translate to potential revenue, and whether revenue increases would offset costs of larger-scale training and inference. In this work, we develop an economic model for characterizing the rational behavior of an LLM training firm by combining scaling laws with microeconomic theory. Under our model of firm behavior, LLM quality can be increased with more parameters and training tokens, leading to more potential adoption by consumers, who each have a quality threshold for using the LLM. On the other hand, additional parameters and training tokens both incur additional costs. We analyze the profit maximization problem for this model under compute-bound and data-bound regimes. In the compute-bound regime, optimal model size and token budget track hardware efficiency $E$ (FLOPs/\$) at a near-linear rate; total training cost then scales sub-quadratically in $E$. Data efficiency improvements incentivize larger models and training expenditure. When we are limited to $D$ data, profit-optimal training expenditure scales as $D^2/E$, i.e, increase with data and decreases with hardware efficiency (as well as data efficiency). Finally, we analyze practical trends in training expenditure: current trends are consistent with our most permissive model variants in the compute-bound regime, but are not profit-optimal in the data-bound regime or assuming hardware advances will stall. Overall, our results provide a theory of profit-optimal LLM training, providing a foundation for engaging critically with industry statements and supporting long-term economic decision making.

Context. Behaviour-Driven Development (BDD) test suites accumulate duplicated step subsequences. Three published refactoring patterns are available (within-file Background, within-repo reusable-scenario invocation, cross-organisational shared higher-level step), but no prior work automates which recurring subsequences are worth extracting or which mechanism applies. Objective. Rank recurring step subsequences ("slices") by refactoring suitability (extraction-worthy), pre-map each to one of the three patterns, and quantify prevalence across the public BDD ecosystem. Method. Every contiguous L-step window (L in [2, 18]) in a 339-repository / 276-upstream-owner Gherkin corpus is keyed by paraphrase-robust cluster identifiers and counted under three scopes. SBERT / UMAP / HDBSCAN clustering recovers paraphrase-equivalent slices. Three authors label a stratified 200-slice pool against a written rubric. An XGBoost extraction-worthy classifier trained under 5-fold cross-validation is compared with a tuned rule baseline and two open-weight Large Language Model (LLM) judges. Results. The miner produces 5,382,249 slices collapsing to 692,020 recurring patterns. Three-author Fleiss' kappa = 0.56 (extraction-worthy) and 0.79 (mechanism). The classifier reaches out-of-fold F1 = 0.891 (95% CI [0.852, 0.927]), outperforming both the rule baseline (F1 = 0.836, p = 0.017) and the better LLM judge (F1 = 0.728, p = 1.5e-4). 75.0%, 59.5%, and 11.7% of scenarios carry a within-file Background, within-repo reusable-scenario, and cross-organisational shared-step candidate, respectively; the figures are stable under a sweep of the classifier decision threshold. Conclusion. Paraphrase-robust subscenario discovery yields a corpus-wide census of BDD refactoring candidates; pipeline, classifier predictions, labelled pool, and rubric are released under Apache-2.0.

Strategic classification studies learning settings in which individuals can modify their features, at a cost, in order to influence the classifier's decision. A central question is how the sample complexity of the induced (strategic) hypothesis class depends on the complexities of the underlying hypothesis class and the cost structure governing feasible manipulations. Prior work has shown that in several natural settings, such as linear classifiers with norm costs, the induced complexity can be controlled. We begin by showing that such guarantees fail in general - even in simple cases: there exist hypothesis classes of VC dimension $1$ on the real line such that, even under the simplest interval neighborhoods, the induced class has infinite VC dimension. Thus, strategic behavior can turn an easy learning problem into a non-learnable one. To overcome this, we introduce structure via a geometric definability assumption: both the hypothesis class and the cost-induced neighborhood relation can be defined by first-order formulas over $\mathbb{R}_{\mathtt{exp}}$. Intuitively, this means that hypotheses and costs can be described using arithmetic operations, exponentiation, logarithms, and comparisons. This captures a broad range of natural classes and cost functions, including $\ell_p$ distances, Wasserstein distance, and information-theoretic divergences. Under this assumption, we prove that learnability is preserved, with sample complexity controlled by the complexity of the defining formulas.

Large foundation models (FMs) are transforming Earth science by integrating heterogeneous multimodal data, such as multi-platform imagery, gridded reanalysis data, diverse geophysical and geochemical observations, and domain-specific text, to support tasks ranging from basic perception to advanced scientific discovery. This paper provides a unified review of Earth science foundation models (Earth FMs) through two complementary dimensions: depth, which traces the evolution of model capabilities from perception to multimodal reasoning and agentic scientific workflows, and breadth, which summarizes their expanding applications across the atmosphere, hydrosphere, lithosphere, biosphere, anthroposphere, and cryosphere, as well as coupled Earth system processes. Using this framework, we review representative multimodal Earth foundation models and compile more than 200 datasets and benchmarks spanning diverse Earth science tasks and modalities. We further discuss key challenges in multimodal data heterogeneity, scientific reliability and continual updating, scalability and sustainability, and the transition from foundation models to agentic and embodied Earth intelligence, and outline future directions toward more integrated, trustworthy, and actionable AI Earth scientists. Overall, this paper offers a structured roadmap for understanding the development of Earth foundation models from both capability depth and application breadth.

Federated learning (FL) in heterogeneous environments remains challenging because client models often differ in both architecture and data distribution. While recent approaches attempt to address this challenge through client clustering and knowledge distillation, simultaneously handling architectural and statistical heterogeneity remains difficult. We introduce COSMOS, a model-agnostic framework that enables server-side personalization using only pseudo-label communication. Clients train local models and predict on the public data; the server clusters clients by prediction similarity, trains a cluster-specific model for each group using its own compute, and distills the resulting models back to clients. We provide the first theoretical analysis showing that distillation from the learned cluster models can yield exponential personalization risk contraction, going beyond the convergence-to-stationarity guarantees typically provided in model-agnostic FL. Experiments across benchmarks demonstrate that COSMOS consistently outperforms all model-agnostic FL baselines while remaining competitive with state-of-the-art personalized FL methods. More broadly, our results highlight personalized server-side learning with pseudo-labels as a promising paradigm for scalable and model-agnostic federated learning in highly heterogeneous environments.

We propose POLAR, a novel radar-guided depth estimation method that introduces polynomial fitting to efficiently transform scaleless depth predictions from pretrained monocular depth estimation (MDE) models into metric depth maps. Unlike existing approaches that rely on complex architectures or expensive sensors, our method is grounded in a fundamental insight: although MDE models often infer reasonable local depth structure within each object or local region, they may misalign these regions relative to one another, making a linear scale and shift (affine) transformation insufficient given three or more of these regions. To address this limitation, we use polynomial coefficients predicted from cheap, ubiquitous radar data to adaptively adjust predictions non-uniformly across depth ranges. In this way, POLAR generalizes beyond affine transformations and is able to correct such misalignments by introducing inflection points. Importantly, our polynomial fitting framework preserves structural consistency through a novel training objective that enforces local monotonicity via first-derivative regularization. POLAR achieves state-of-the-art performance across three datasets, outperforming existing methods by an average of 24.9% in MAE and 33.2% in RMSE, while also achieving state-of-the-art efficiency in terms of latency and computational cost.

Recent work on text diffusion models offers a promising alternative to autoregressive generation, but controlling their safety remains underexplored. Existing safety approaches are geared toward autoregressive models and typically rely on post-hoc filtering or inference-time interventions. These are inadequate for effectively addressing safety risks in text diffusion models. We propose the Safety-Aware Denoiser (SAD), a safety-guidance framework in text diffusion models. The SAD modifies the iterative denoising process such that the text sample at the final denoising step is steered toward provably safe regions of the text space. This inference-time method can integrate safety constraints into the denoiser, avoiding computationally expensive retraining of the underlying diffusion model and enabling flexible, lightweight safety guidance. We evaluate the safety of the generated text using the SAD, with respect to hazard taxonomy, memorization, and jailbreak. Experimental results show that SAD substantially reduces unsafe generations while preserving generation quality, diversity, and fluency, outperforming existing methods. These results demonstrate that our safety guidance during denoising provides an effective and scalable mechanism for enforcing safety in text diffusion models.

Precision oncology is currently limited by the small-N, large-P paradox, where high-dimensional genomic data is abundant but pharmacological response samples are sparse. While deep learning achieves predictive accuracy, it frequently fails to provide the mechanistic clarity required for clinical adoption. We present the Contextual Invertible World Model (CIWM), a Neuro-Symbolic Agentic Framework that bridges this gap by integrating a quantitative machine learning emulator with a Large Language Model reasoning layer. Utilising a stringently curated, high-fidelity data engineering pipeline on the Sanger GDSC dataset (\( N=83 \)), we isolate true biological signals from in vitro artifacts to establish a rigorous baseline predictive correlation for complex transcriptomics (\( r=0.268 \)). Through Inverse Reasoning, we perform in silico CRISPR perturbations across the colorectal landscape. The framework autonomously overturns classical mechanistic assumptions, identifying a hierarchical dominance of mutant KRAS over the APC/Wnt-axis in driving 5-fluorouracil resistance (\( \Delta=-0.0469 \)) via a "KRAS Shield" mapped to MAPK/PI3K networks. Furthermore, the agentic layer identified a "PIK3CA Paradox", revealing that repairing PIK3CA inadvertently increases chemoresistance (\( \Delta=+0.0085 \)) by triggering a compensatory feedback loop that hyperactivates the dominant MAPK survival pathway.

Existing benchmarks for Vision-Language Models (VLMs) primarily evaluate spatio-temporal understanding on simple single-action videos, closed attribute sets and restricted entity types, failing to capture the freeform, multi-action interactions between diverse entities which characterize real-world video understanding. Furthermore, the lack of a systematic framework for analyzing model failures across complementary spatio-temporal axes hinders comprehensive evaluation. To address these gaps, we introduce VISTA, a Video Interaction Spatio-Temporal Analysis benchmark designed for open-set, multi-entity and multi-action spatio-temporal understanding in VLMs. VISTA decomposes videos into interpretable entities, their associated actions, and relational dynamics, enabling multi-axis diagnostics and unified assessment of relational, spatial, and temporal understanding. Our benchmark integrates multiple datasets into a single interaction-aware taxonomy and comprises ~12K curated video-query pairs spanning diverse scenes and complexities. We systematically evaluate 11 state-of-the-art VLMs on VISTA, and break down aggregate performance across our taxonomy to reveal shortcomings and pronounced spatio-temporal biases obscured by traditional metrics. By providing detailed, taxonomy-driven diagnostics on a challenging dataset, VISTA offers a nuanced framework to guide advances in model design, pretraining strategies, and evaluation protocols. Overall, VISTA is the first, large-scale, interaction-aware diagnostic benchmark for spatio-temporal understanding in VLMs.

Retrieval-Augmented Generation (RAG) extends large language models (LLMs) beyond parametric knowledge, yet it is unclear when iterative retrieval-reasoning loops meaningfully outperform static RAG, particularly in scientific domains with multi-hop reasoning, sparse domain knowledge, and heterogeneous evidence. We provide the first controlled, mechanism-level diagnostic study of whether synchronized iterative retrieval and reasoning can surpass an idealized static upper bound (Gold Context) RAG. We benchmark eleven state-of-the-art LLMs under three regimes: (i) No Context, measuring reliance on parametric memory; (ii) Gold Context, where all oracle evidence is supplied at once; and (iii) Iterative RAG, a training-free controller that alternates retrieval, hypothesis refinement, and evidence-aware stopping. Using the chemistry-focused ChemKGMultiHopQA dataset, we isolate questions requiring genuine retrieval and analyze behavior with diagnostics spanning retrieval coverage gaps, anchor-carry drop, query quality, composition fidelity, and control calibration. Across models, Iterative RAG consistently outperforms Gold Context, with gains up to 25.6 percentage points, especially for non-reasoning fine-tuned models. Staged retrieval reduces late-hop failures, mitigates context overload, and enables dynamic correction of early hypothesis drift, but remaining failure modes include incomplete hop coverage, distractor latch trajectories, early stopping miscalibration, and high composition failure rates even with perfect retrieval. Overall, staged retrieval is often more influential than the mere presence of ideal evidence; we provide practical guidance for deploying and diagnosing RAG systems in specialized scientific settings and a foundation for more reliable, controllable iterative retrieval-reasoning frameworks.

Deep neural networks achieve impressive results across diverse applications, yet their overconfidence on unseen inputs necessitates reliable epistemic uncertainty modeling. Existing methods for uncertainty modeling face a fundamental dilemma: Bayesian approaches provide principled estimates but remain computationally prohibitive, while efficient second-order predictors lack rigorous connections between their specific objectives and epistemic uncertainty quantification. To resolve this dilemma, we introduce Dirichlet-approximated possibilistic posterior predictions (DAPPr), a principled framework grounded in possibility theory. We define a possibilistic posterior over parameters, project it to the prediction space via supremum operators, and approximate the projected posterior using learnable Dirichlet possibility functions. This projection-and-approximation strategy yields a simple training objective with closed-form solutions. Despite its simplicity, extensive experiments across diverse benchmarks show that DAPPr achieves competitive or superior uncertainty quantification performance over state-of-the-art second-order predictors while maintaining both principled derivation and computational efficiency. Code is available at this https URL.

Online conformal prediction must balance fast adaptation to distribution shift against stable coverage: feedback-driven methods react quickly but become volatile, while strongly discounted Bayesian methods lag and inflate intervals at tight coverage. We introduce \textbf{State-Adaptive Bayesian Conformal Prediction (SA-BCP)}, which forms the predictive quantile as a gated convex combination of long-term temporal inertia and local spatial evidence from a kernel density estimate, controlled by a single interpretable evidence threshold $K$. We establish three results: (i) asymptotic marginal validity of the resulting intervals; (ii) a closed-form expression for the MSE-optimal threshold, $K^*_{\mathrm{MSE}}=\alpha(1-\alpha)/M^{\mathcal{T}}$, trading the coverage-indicator (Bernoulli) variance against the temporal structural bias $M^{\mathcal{T}}$; and (iii) a rolling-origin procedure for selecting $K$ online -- consistent under stationarity, with $O(\sqrt{T\log N})$ regret against the best fixed $K$ and, for a segmented variant, a sublinear dynamic-regret bound under bounded drift. Across four financial-volatility and weather datasets, three target coverage levels, and eight baselines (including the strongest recent conditional-quantile methods, SPCI and KOWCPI), SA-BCP attains at-or-above-nominal coverage in most settings while producing substantially sharper intervals -- up to roughly $3\times$ lower Winkler score than discounted Bayesian CP at the tightest coverage -- and a coverage-matched audit confirms these efficiency gains are not an artifact of under-coverage. We disclose one principal limitation: a volatility-specialized conformal-GARCH competitor remains more efficient on its home volatility-base series, though it does not transfer across domains.

Recent large language models (LLMs) have achieved impressive reasoning milestones but continue to struggle with high computational costs, logical inconsistencies, and sharp performance degradation on high-complexity problems. While neuro-symbolic methods attempt to mitigate these issues by coupling LLMs with symbolic reasoners, existing approaches typically rely on monotonic logics (e.g., SMT) that cannot represent defeasible reasoning -- essential components of human cognition. We present "LLM+ASP," a framework that translates natural language into Answer Set Programming (ASP), a nonmonotonic formalism based on stable model semantics. Unlike prior "LLM+ASP" approaches that require manually authored knowledge modules, domain-specific prompts, or evaluation restricted to single problem classes, our framework operates without any per-task engineering and applies uniformly across diverse reasoning tasks. Our system utilizes an automated self-correction loop where structured feedback from the ASP solver enables iterative refinement. Evaluating across six diverse benchmarks, we demonstrate that: (1) stable model semantics allow LLMs to naturally express default rules and exceptions, outperforming SMT-based alternatives by significant margins on nonmonotonic tasks; (2) iterative self-correction is the primary driver of performance, effectively replacing the need for handcrafted domain knowledge; (3) compact in-context reference guides substantially outperform verbose documentation, revealing a "context rot" phenomenon where excessive context hinders constraint adherence.