GARCH-LSTM hybrid models have become increasingly popular in the volatility forecasting literature. To date, these models have almost exclusively focused on point estimation and accuracy measures such as RMSE and MAPE, while failing to assess the uncertainty of these point estimates. We approach this gap in two ways. First, we extend the existing GARCH-LSTM model by modeling the conditional mean with an ARMA structure. Second, we develop a forward bootstrap algorithm to build multi-step-ahead prediction intervals for future realized volatility. In order to address the notorious undercoverage of prediction intervals, we implement two solutions. One draws bootstrap innovations from the empirical residual distribution to mimic the fat-tail property of financial series. The other utilizes principles from conformal prediction to systematically extend the prediction interval length. We evaluate the methods in simulations with heavy-tailed and time-varying GARCH processes and in rolling-window forecasts of four U.S. stock indices from 2020 to 2025. The ARMA-based hybrid gives the lowest point-forecast errors in most settings, and combining both remedies with the mean-based hybrid yields intervals closest to the nominal level among competing methods.
We study the problem of agnostic PAC learning under deterministic labels, where the examples are labeled by an arbitrary deterministic function, which need not belong to the concept class $C$. This model captures misspecification without stochastic label noise and lies between the realizable and fully agnostic settings. Prior work showed that, when $C$ has finite diameter $K$ and VC dimension $d$, deterministic labels can improve the sample complexity upper bound from the classical agnostic guarantee $O(d/\varepsilon^2)$ to $\widetilde{O}(K/\varepsilon)$, but left a large gap between upper and lower bounds. Our main result sharpens this picture. We prove that every class of VC dimension $d$ and finite diameter $K$ is learnable with sample complexity $\widetilde{O}\left( \frac{K^{1/3} \, d^{2/3}}{\varepsilon} \right)$, and we construct classes for which this bound is tight up to logarithmic factors. We also study the online analogue of this problem and show that the minimax regret is again partly controlled by the diameter of the class. Technically, our upper bounds are obtained via memorization-based learning algorithms that exploit the deterministic label structure.
Cultural adaptation in dialogue systems refers to the ability to generate responses that conform to linguistic norms and social expectations across different cultures. However, existing dialogue generation approaches pay limited attention to the role of personality in shaping cultural expression and communication style. We therefore propose HiCAP, a novel framework that integrates personality traits into culturally adaptive dialogue generation. HiCAP employs a personality-conditioned control mechanism that hierarchically modulates response style by first aligning it with the cultural tone and then refining linguistic expression and content according to individual personality attributes. To embed these factors, the model combines supervised fine-tuning and instruction-based tuning, supported by dynamic instruction templates and multi-agent demonstrations to capture diverse combinations of culture and personality. Experiments on CulturalPersonas and CPED show that HiCAP achieves the strongest overall performance across various dialogue generation metrics, demonstrating effective cultural adaptability and personality modelling.
Feedback controllers for flows are expensive to design, since each candidate must be evaluated with a high-fidelity solver. Differentiable surrogates reduce that cost, but they also introduce a failure mode that prediction error cannot detect. Policy optimization uses the derivative of the objective with respect to the action, which we call the action gradient and which can be computed either through the surrogate or through the high-fidelity solver. Optimization through the surrogate uses only its action gradient, and a surrogate can follow the flow trajectory closely while misestimating this derivative. The surrogate action gradient is compared against a central finite-difference estimate from the high-fidelity solver on held-out states, under acceptance criteria fixed before the comparison, and we call that comparison an audit. A low-dimensional feedback law is then fitted to the surrogate action gradient when the surrogate passes the audit, and to the solver estimate when it does not. This law initializes a neural policy, and a quadratic penalty holds the policy near it during optimization. The proposed approach is validated on four test cases. The resulting controllers outperform or match a model-free baseline, while using 3 to 156 times fewer high-fidelity solver interactions across the four test cases. One surrogate predicts the objective at correlation 0.9996 but its action gradient at only 0.459, so prediction accuracy does not identify a usable surrogate. Another surrogate, whose action gradient agrees with the solver estimate at correlation 0.996, makes the objective 0.52% worse when the policy is optimized through it without the anchor penalty, so an accurate gradient alone does not provide a working controller. Finally, a surrogate that passes the audit at one fixed action fails it around the actions the deployed controller applies, so a pass is local.
Thread density, the threads per inch along and across a woven fabric, is a basic specification that buyers check whenever fabric is sold. Without laboratory instruments, inspectors count it slowly, by eye, through a pick glass, a folding 10$\times$ magnifier with a one-inch window. A smartphone photo through that glass could make the check faster and leave a record. We judge such a photo by what it is used for, as in task-based image quality: it is good if the thread count from it is within $\pm$5% of the true count, a common tolerance in fabric specifications. We introduce PickGlass-TPI: 76 factory scenes of 44 fabric patches, each counted in an open annotation tool by a textile-trained annotator, whose counts are the reference. On it we benchmark frontier and open vision-language models, CNNs, counting specialists and a training-free spectral counter (PG-Lattice). No method agrees with the reference within the tolerance. The best methods are off by about a quarter: a small open model that we tuned only on synthetic images (Molmo2-4B, 25%) is level with the best frontier model, GPT-6.1-sol (27% when asked directly, 25% given four magnified close-ups of the window). Both err less than a non-expert annotator (37%), but only the tuned model also beats, on both error measures, a baseline that guesses each scene's density as the average of the others. PG-Lattice most often repeats its count within 5% across two photos of the same patch, but its misses, like GPT-6.1-sol's errors, are often a factor of two. Even the reference repeats a count across such re-shoots only to a median 9%, so one photo cannot yet be shown to meet $\pm$5%; the image quality that matters is reliability across re-shoots. Generic no-reference quality scores chosen before labeling do not rank PG-Lattice's error, and generative enhancement raises perceptual scores but does not reliably improve the count.
Machine learning-based resource management (RM) models degrade under out-of-distribution (OOD) inputs at runtime, yet existing approaches in the literature rely on a single global error margin that is overly conservative and fails to distinguish reliablepredictions from catastrophic failures. We propose Lightweight Runtime EXplainability (LEX), the first post-hoc XAI framework for input-specific runtime confidence estimation in ML-based RM. LEX clusters the embeddings of the training input space using a Gaussian Mixture Model and associates per-cluster prediction error bounds to input samples, enabling fine-grained runtime confidence signals without retraining. Evaluated on a state-of-the-art LSTM-based IPS forecasting model on Intel Alder Lake, across 34 benchmark applications, LEX achieves ~94% containment on unseen workloads and produces ~17% tighter per-input confidence bounds than global baselines. With its negligible runtime overhead, LEX advances the field by providing explainable and reliable input-specific confidence signals of ML predictions, thereby making ML-based resource management resilient to OOD at runtime.
We study the problem of cross-sample modality attribution, which involves measuring how one modality in one sample affects the learning outcome in other multimodal samples. This problem is crucial in a multimodal system because a data modality (such as a text description) may be redundant for one sample (a self-explanatory image), but the same data modality would otherwise be uniquely informative for another sample, or informative when combined with other modalities. Prior work focuses on influence estimation at the level of one sample, and it remains an open question how one should capture the interactions across both samples and modalities. To capture scenarios like the above, we propose structured kernel surrogate models, which are surrogate mappings from both samples and modalities to the prediction outcome when they are trained together. We design an algorithm, SKS, which applies kernel ridge regression with a structured, compositional kernel on the selection states of individual samples and modalities. Specifically, the kernel function multiplies kernel-based similarity scores between every pair of data modalities for one sample and a pair of two samples. Additionally, we design a gradient-based estimation algorithm to efficiently estimate the structured kernel surrogate model. We evaluate our approach on two multimodal benchmarks, spanning a range of data modalities including satellite images, road networks, text, audio, and tabular data. We find that SKS achieves $57.8$% higher correlation with the true prediction outcomes than existing methods. In downstream data selection, SKS improves the average accuracy by $1.93$% over baselines. Models with data selected by SKS have lower Hessian traces, indicating a flatter loss surface. Using SKS to select satellite images identifies road segments, where accident scenes commonly occur, as opposed to background images.
LLM-based agents can reason, use tools and collaborate. However, coordinating increasingly complex agentic applications becomes difficult to understand, monitor and improve. Decisions distributed across prompts, tool calls and application logic make workflows implicit. We introduce **Recursive Capability Composition**, enabling agents to execute business processes and expose them as reusable Agentic Tools. This paper presents a BPMN-based approach to Agentic Workflows, combining explicit orchestration with autonomous reasoning. SemTalk BPMN models are transformed into executable workflows using a LangGraph/Pregel runtime. Agentic Tasks introduce LLM reasoning at selected activities, while conventional BPMN controls process flows, deterministic decisions, system interactions and human approvals. Agentic Participants provide reusable interfaces for agent behavior or other executable processes. BPMN orchestration makes multi-agent systems more understandable and manageable. Explicit process models provide developers and operators with a shared view of workflow structures and execution progress, facilitating debugging, monitoring and optimization. Enterprise agents also need organization-specific context. Process Models and Knowledge Graphs describe activities, participants, information objects, decisions and relationships between processes. They provide structured knowledge of how an organization operates, supporting informed agent decisions and actions. Our contributions are threefold: 1. **Model-Driven Execution:** Transforming BPMN models into executable Agentic Workflows. 2. **Recursive Capability Composition:** Exposing agents and executable business processes as reusable tools for other agents. 3. **Enterprise Context:** Using Process Models and Ontologies to provide organization-specific knowledge for agent reasoning.
Accurate and timely identification of road closures is critical for large-scale navigation and logistics systems, yet remains challenging due to the need to jointly reason over visual evidence, spatial context, and road network topology. Existing approaches either rely on manual map updates or focus on detecting closure-related objects without resolving which road segment is affected. In this work, we introduce a unified multimodal and multitask framework that jointly performs road closure classification (no, partial, complete) and road segment association within a single learning architecture. Our model integrates graph-structured map data, satellite imagery augmented with projected street-level detections, and street-view images using modality-specific encoders and cross-modal transformer attention, enabling explicit reasoning over visual cues and network topology. Evaluated on nearly 20K real-world scenarios collected from a large-scale commercial vehicle fleet, our approach achieves 72.7% closure classification accuracy and 60% road segment association accuracy, outperforming frontier vision-language models by over 28 percentage points on closure classification while being over 4,000x smaller and 100x faster at inference. Extensive ablations demonstrate the importance of cross-modal attention, edge-aware pooling, and graph message passing for accurate geospatial reasoning. This work establishes the first benchmark for joint road closure classification and segment-level attribution, and shows that task-specific architectural inductive biases substantially outperform general-purpose multimodal models for structured geospatial understanding.
Background: Geographic atrophy (GA) is an advanced form of age-related macular degeneration (AMD). It causes irreversible vision loss due to progressive degeneration of the retinal pigment epithelium (RPE) and photoreceptors. Accurate segmentation of GA in en-face image is crucial for early diagnosis, monitoring disease progression, and developing personalized treatment strategies. However, existing deep learning approaches are typically evaluated using single architectures and single-device datasets, limiting insight into relative performance and cross-device robustness. Methods: We trained and systematically compared five deep learning architectures for GA segmentation: Nested U-Net, ResU-Net, nnUnet, Swin-Unet and MedSegDiff model, together with a thresholding based classical baseline. The models were trained on a dataset of en-face OCT images derived from 269 OCT volumes from two different devices, following a three-step preprocessing technique involving normalization and inpainting to enhance the quality of the en-face images. Results: On the fixed held-out validation set comprising OCT scans from both Cirrus (R) (Carl Zeiss Meditec, Inc., USA) and Spectralis (R) (Heidelberg Engineering, Germany) devices, nnUnet achieved the highest mean Dice similarity coefficient (DSC) of 0.8882 +/- 0.0033. The other models yielded the following Dice scores: Nested U-Net (0.7932 +/- 0.0423), ResU-Net (0.8453 +/- 0.0074), Swin-Unet (0.6886 +/- 0.0294), and Diffusion model (0.7959 +/- 0.0242). Conclusions: nnUnet framework, demonstrated most reliable performance for GA segmentation in en-face OCT under limited data conditions. It offers the best balance of accuracy, stability, and efficiency for GA quantification.
Accurate cancer survival prediction remains challenging due to tumor heterogeneity. Although multimodal survival models integrate clinical and molecular data, they often overlook evolutionary information encoded in genomic alterations. To address this gap, we developed eTime, a 38-dimensional evolutionary representation comprising eight feature groups that characterize clonal and subclonal mutation burden, mutational spectra, functional consequences, copy-number alterations, and anoikis-related genomic changes. We further developed EvoSurv, a multimodal survival framework integrating eTime with clinical variables and RNA expression. Across six TCGA cancer cohorts comprising 2,782 patients, EvoSurv achieved the highest mean concordance index (C-index) in all six cancers, averaging 0.663 and outperforming three state-of-the-art survival prediction methods. More importantly, we assessed whether evolutionary features provide additional prognostic information beyond clinical and transcriptomic predictors. eTime remained significantly associated with survival in head and neck squamous cell carcinoma (HNSC) and liver hepatocellular carcinoma (LIHC) after adjustment for clinical variables and RNA-derived risk scores. Feature-group analyses further revealed cancer-specific prognostic patterns. In HNSC, T>G substitutions were associated with increased mortality risk after clinical and molecular adjustment, including in HPV-negative patients, with supporting evidence from an independent CPTAC-HNSCC cohort. Overall, these findings demonstrate that eTime captures survival-associated evolutionary information and highlight its potential to provide prognostic value beyond clinical and transcriptomic predictors.
Expert parallelism (EP) shards the experts of a Mixture-of-Experts (MoE) layer across GPUs, so every MoE layer requires two all-to-all communications, namely dispatch and combine. Compared with training, inference processes relatively few tokens per GPU at a time, and at small batch sizes fixed costs that do not grow with the number of tokens, rather than data movement, dominate the latency. We present NVLinkOneSided, a one-sided EP all-to-all for GPUs within a single NVLink domain. Each source rank pushes its tokens, without first exchanging metadata, into rank-major buffers on the target GPUs, and the grouped GEMMs read them in place instead of from a per-expert copy. We provide two protocols for the synchronization that the communication requires, namely Fence and Sentinel, for high-throughput and low-latency scenarios respectively. On Blackwell GPUs, NVLinkOneSided completes dispatch and combine 1.2-3.0$\times$ faster than DeepEP, with the largest speedups at small batch sizes. The resulting MoE layer is up to 2.4$\times$ faster than DeepEP with DeepGEMM. It even outperforms the fused DeepGEMM MegaMoE at up to 128 tokens per GPU, making it still competitive for decoding. We expose NVLinkOneSided through a modular interface, so that it is agnostic to the activation quantization format, decoupled from the MoE compute backend and easy to integrate into inference frameworks.
Ranking large language models (LLMs) through human evaluations is essential for downstream decision-making. Because human feedback is expensive, evaluation platforms often collect it adaptively, prioritizing informative model comparisons using previously observed feedback. To further improve precision with limited human feedback, evaluators can draw on abundant unrated responses and relatively inexpensive LLM judges; however, judge bias and adaptive dependence challenge standard i.i.d. inference frameworks. We propose Prediction-Powered Ranking (PP-Rank), a framework that combines LLM-judge predictions with human feedback to improve precision while providing valid inference under adaptive sampling. Data-driven prediction weights minimize the asymptotic covariance trace within the proposed class. In theory, we establish asymptotic normality for general evaluation targets and simultaneous ranking inference. A joint martingale analysis handles dependence between adaptive human feedback and auxiliary data. We further derive a rule for reducing exploration that prioritizes lower-cost comparisons while matching a conservative variance bound, and characterize allocation limits for regular $\sqrt{T}$-rate estimation of pairwise means. Semi-synthetic experiments on UltraFeedback-Extended show improvements over human-only baselines in score estimation, confidence interval width, and ranking certification. Synthetic experiments demonstrate robustness to judge quality and adaptive allocation, while showing that PP-Rank can substantially reduce evaluation costs without sacrificing precision.
Domain-specific languages (DSLs) are widely used in enterprise software, giving a compact interface to complex tasks, from data access to infrastructure automation. Platforms increasingly extend them to agents, where a developer declares the topics, permitted actions, and guardrails of an agent rather than writing general-purpose code. Large language models (LLMs) write general-purpose code well, but that ability has been measured almost entirely on languages abundant in public pretraining data, and enterprise DSLs are neither abundant nor public. We introduce AgentScript-Eval, an execution-graded benchmark of 100 real tasks for AgentScript, the declarative DSL that Salesforce uses in Agentforce to build conversational agents. Tasks span authoring an agent from a specification, debugging a broken definition, and optimizing a working one. Evaluating five models zero-shot, we find that the regime is hard: the best model reaches only 25.0% pass@3, and every model scores 0% on authoring from scratch. We further find that model scale and price are weak predictors of success -- a 31B open-weight model attains the highest compilation rate of any model tested (48.0%) and matches the strongest frontier model on partial-credit accuracy at the lowest token cost. Adding an agentic compile-and-edit loop is the single largest lever, roughly doubling partial-credit accuracy. Beyond that gain, the wording of domain guidance matters more than its packaging. The same skill pack scores 28% as plain documentation but only 14% as a live plugin. Rewriting one skill inside that plugin then raises it to 37%, our best result, with catalog and tooling untouched. Larger models did not reliably outperform smaller ones; what mattered was the harness surrounding the model and the guidance supplied to it.
Deep reinforcement learning (DRL) methods have been the dominant paradigm in continuous control tasks due to their efficacy in function approximation, while interpretable evolutionary methods have lagged largely behind for the opposite reason. In this work, we introduce Symbolic Imitation Learning, a framework for distilling an oracle policy through an offline dataset into a decision list with linear model motion primitives. We show that this, in conjunction, with online Augmented-Random Search (ARS) tuning of the constants and motion primitives approximates the performance of the oracle across 20 benchmark RL tasks and exceeds the oracle in 11 of those 20 tasks while simultaneously producing parsimonious, interpretable policies that domain experts can better understand.
Image geolocalization predicts the location at which a photo was taken. Accurate geolocalization requires discriminating between both physically and semantically adjacent areas on Earth. However, prior diffusion-based approaches predict locations by generating coordinates, which are semantically unstable over space. We avoid this limitation by learning geolocalization behavior in the pretrained representation spaces of location encoders, which provide both positional and semantic understanding. To make this space usable for coordinate prediction, we introduce a Location Autoencoder (LAE), which pairs a frozen location encoder with a learned decoder mapping embeddings back to coordinates with kilometer-level accuracy worldwide. In the LAE space, we train DIG, a latent diffusion model that transports noise to location embeddings conditioned on image features. DIG consistently outperforms coordinate-based generative models and sets a new state of the art for parametric geolocalization on three benchmarks: OSV-5M, Im2GPS3k, and YFCC4k. Notably, DIG is the first generative model to outperform retrieval-based methods, and further improves retrieve-and-rerank systems when used as a prior. Code and models will be released.
Generative pipelines can now turn every product page of an e-commerce catalog into a short video ad, but deciding whether such an ad is usable is not one good-or-bad call. We describe how we operationalized the review and editing of these ads so that both can keep pace with generation: reviewers go through each ad scene by scene and answer six questions, on the product, scene quality, pronunciation, script, sound and captions; judges built from vision-language, speech and language models file each finding under the matching question, and reviewers accept or reject each finding and add what the judges missed. Each confirmed issue then opens an edit tool matched to it. The questions and the routing from a finding to its fix do not depend on the models that generate the ad. Across 120 ads from our pipeline, reviewed with judge findings on screen, reviewers marked the product wrong on 60% of scenes, and script, caption or sound problems on under 10% each. On 60 of these ads, they marked the product wrong on 58% of scenes, far more than the 26% marked by two reviewers who later saw the ads with the judges hidden; one reviewer who saw the ads both ways marked 57% with judge findings and 19% without. On 120 blind-labeled scenes, a product judge that watches the clip with the catalog photos reached AUC 0.84, while video-quality and photo-similarity scores were close to chance. We tune the judges for recall: the product judge catches nearly all defects but also flags many clean scenes, so reviewers decide; together, the judges widen what reviewers mark and reviewers add what the judges miss. Without judge findings, one reviewer's marking faded over a session. We release a review-portal template and our review data (120 ads and 562 scenes with judge findings and human marks, plus 120 blind-labeled scenes with baseline scores) at https://github.com/anoubhav/ad-review-and-edit.
Graphical user interface (GUI) grounding maps a screenshot and a natural-language instruction to the screen location where an agent should act. Zoomed views can reveal small targets but may lose the context needed to identify them. We introduce GroundCascade, a training-free controller that uses agreement between a grounder's full-screen and zoomed predictions to decide when to consult a complementary model. It selects between zoomed predictions while preserving corroborated full-screen locations. An optional Finalizer predicts from an unmarked joint crop when disagreement remains (Full); omitting it yields the two-stage Selective variant. GroundCascade reaches 85.33% on ScreenSpot-Pro with Indeed-UI-32B/8B/32B and 88.43% on the visible-target OSWorld-G-Refined subset with Qwen3.6-27B/3.8-27B/3.8-27B, gains of 2.66 and 2.35 percentage points over the stronger standalone two-pass grounder. On UI-Vision, GroundCascade reaches 72.20% category-balanced accuracy, with Selective slightly higher at 72.28%. Across these three evaluations, GroundCascade in Full mode uses 2.42-2.73 calls per request, 31-39% fewer than unconditionally acquiring both models' two-pass predictions. At matched second-model call budgets, selective acquisition gains 1.52-2.56 points over standalone Two-pass, compared with 0.22-0.69 points from stratified random acquisition. GroundCascade leads the official ScreenSpot-Pro leaderboard and exceeds reported GUI-specialist comparisons on OSWorld-G-Refined and UI-Vision, subject to differences in external evaluation protocols.
Text-to-image diffusion models can render a photorealistic analog clock but cannot control the time it displays. Across four pretrained models, fewer than 1% of generated clocks place both hands within 6 degrees of the requested time; the models instead collapse onto visually dominant configurations such as the canonical 10:10 pose. The same failure appears when a specific fraction of a circular object is requested, where models return complete or conventionally sliced objects. We observe that these two failures share a geometric structure: both target quantities can be parameterized by angular variables. We introduce a geometric token encoding framework that conditions generation directly on these angles. Each angular component is encoded as a periodic sine-cosine feature and projected into a role-specific token embedding that is injected into the prompt before a frozen text encoder. To spatially ground these tokens without requiring localization input at inference, we introduce Stochastic Localization Relaxation (SLR), which stochastically relaxes attention supervision from component-level to object-level to unconstrained localization during training. For clocks, we additionally introduce a hand-swap preference objective that discourages assigning the correct angles to the wrong hand roles. On analog clocks our method reaches 63-78% Acc@6 degrees across evaluation methods, compared with below 1% for the baselines. On circular-fraction generation, it achieves 89.5% Acc@0.10, while maintaining 88.8% on unseen object categories and 84.9% on unseen target values, compared with roughly 20% for pretrained baselines.
AI-generated radiology reports are usually evaluated by comparison with a radiologist-authored reference, leaving no way to score them when no reference exists. We introduce RadRef, a reference-free metric that scores report claims against image-derived beliefs over a clinical finding vocabulary, weighted by relevance to the clinical indication. On ReXVal, RadRef achieves 73.1% agreement on non-tied same-radiograph report pairs with radiologist rankings, comparable to leading reference-based metrics overall. It is also competitive on false findings and has its clearest relative result on omissions (missed findings). We further evaluate 100 model-generated and 100 radiologist-authored reports from 100 MIMIC-CXR studies, all independently annotated by two board-certified radiologists. On omissions, RadRef attains Kendall's tau-b = 0.295, against 0.386 between the two readers themselves, and no compared metric differs significantly from RadRef on this target. The same image-grounded representation provides a false-finding readout, RadRef-FF, with tau-b = 0.32 on ReXVal and 0.29 in the reader study, without requiring a reference. Controlled perturbations confirm that RadRef depends on the radiograph, report, and clinical indication. RadRef enables image-grounded evaluation before a radiologist-authored reference exists, supporting quality monitoring of report-generation systems in clinical settings, and complements reference-based metrics when one is available.