Papers

Vision-Language-Action (VLA) models have emerged as a powerful paradigm for end-to-end autonomous driving by jointly integrating perception, reasoning, and decision making within a unified multimodal framework. However, most existing VLA models formulate end-to-end autonomous driving as a visual question answering task, leading to unreliable and less interpretable decision reasoning. In addition, they fail to establish effective multi-modal interaction across heterogeneous sensors, thereby limiting robust scene perception and reliable driving reasoning in long-tail driving scenarios. To this end, we propose a robust VLA-based end-to-end autonomous driving system that combines multi-modality interaction with multi-trajectory planning and optimization, enabling more reliable, interpretable, and safer driving decisions. Our method comprises three core components: (1) Affinity-Guided Optimal Transport for main-auxiliary modality two-way interaction; (2) Distribution-Consistent Modality Transfer for heterogeneous modality distribution transfer and cross-modal interaction; (3) Multi-modal Multi-Trajectory Planning along with Perception-Oriented Trajectory Refinement for better driving decisions to long-tail driving scenarios. Experimental results in open-loop and closed-loop datasets demonstrate improvements in safety long-horizon driving reasoning and road scene perception over existing driving systems, highlighting the ability of our mutli-modality interaction and multi-trajectory planning and optimization for scalable VLA-based systems.

Automatic Medical Coding (AMC), which assigns standardized International Classification of Diseases (ICD) codes to clinical notes, is essential for medical reimbursement, quality reporting, and clinical research. Existing pre-trained language model (PLM)-based methods typically formulate AMC as an extreme multi-label classification problem over a predefined code set, while recent large language model (LLM)-based approaches instead frame it as generation or multi-step reasoning. However, key challenges remain, including the extreme length of clinical notes that hinders effective interpretation, the vast ICD label space, and complex coding rules that are not explicitly captured by LLMs. In this work, we propose Knowledge-Guided Reasoning over Clinical Evidence with LLMs (KREL), a framework that leverages LLMs for clinical text understanding and reasoning while integrating external ICD coding guidelines as structured knowledge. This design enables tight coupling between domain knowledge and LLM reasoning, reducing hallucinations and improving compliance with coding standards. Experiments on benchmark datasets show that KREL consistently outperforms strong PLM-based and state-of-the-art LLM-based baselines.

Real-world image search queries are multimodal and compositional: “find this shirt in pink” specifies an entity to retain, an attribute to modify, and context to ignore. Yet existing re-rankers either compress such multifaceted relevance into an opaque embedding or rely on free-form chain-of-thought that easily omits or hallucinates fine-grained constraints. Drawing on rubric- and checklist-based evaluation from NLP, we recast multimodal image re-ranking as a semantic constraint satisfaction problem and propose EviRank, which parses any query - text-only, image-only, or composed - into a unified evidence package: typed criteria across six semantic slots (e.g., entities, attributes, relations), each labelled required, forbidden, or ignorable. Re-ranking then reduces to evidence-conditioned verification, combining deterministic rubric scoring and evidence-grounded listwise comparison in a single training-free procedure. The explicit evidence can further serve as structured supervision for optionally distilling a lightweight student. Across five benchmarks spanning text-to-image, image-to-image, and composed image retrieval, EviRank achieves state-of-the-art performance, and the distilled student preserves over 90% of the teacher's capability at substantially lower cost.

Unsourced random access (URA) supports massive sporadic connectivity but remains limited by multiuser interference, distributed channel uncertainty, and local fading. This paper proposes a fluid-antenna-assisted distributed joint decoder for cell-free massive MIMO URA. Single-radio-frequencychain access points sequentially sound correlated ports; the central processor then performs correlation-aware expectationmaximization approximate message passing, selects one common payload port per access point, and combines MIMO iterative Gaussian approximation with full-packet channel re-estimation and successive interference cancellation. Fixed-port ablations confirm the complementary benefits of distributed reception, reestimation, and iterative cancellation, while fluid-antenna simulations show improved average and lower-tail received power over fixed-port reception. The framework provides a feasible integration of fluid antennas and cell-free URA and motivates end-to-end evaluation under pilot collisions and switching overhead.

Face recognition systems (FRSs) are increasingly deployed in critical real-world services for authentication, such as banking applications and airport identity checks, necessitating stringent security configurations. Consequently, the security vulnerabilities of FRSs have garnered significant attention. While existing studies have extensively explored FRS security, prior analyses have primarily focused on medium-security threshold settings, which are not directly applicable to FRSs operating under high-security constraints. In this paper, we propose the first successful impersonation attack against FRSs under high-security threshold settings. Among various threat models, we focus on a practical and challenging scenario: score-based impersonation attacks under strict rate limits. To precisely evaluate the feasibility of such attacks, we provide a principled mathematical analysis characterizing the gaps in each stage of the attack pipeline. Our method significantly enhances impersonation capabilities in score-based attacks, even under elevated decision thresholds. On the LFW benchmark, with a budget of only 100 confidence score queries per identity, our attack achieves an impersonation success rate exceeding 92% against Amazon Rekognition at a confidence score threshold of 99-recommended setting for law enforcement scenarios. We further observe consistently robust performance across multiple open-source FRSs evaluated at similarly stringent decision thresholds.

Modern generators faithfully model macroscopic semantics, producing synthetic images that appear highly realistic. Consequently, decisive forensic cues reside in subtle non-semantic visual discrepancies. To reveal these cues, we revisit AIGI detection from a geometric perspective and identify an architecture-agnostic signature. Specifically, modern generators exhibit low-rank collapse (i.e., rank degeneracy) in the semantic-residual orthogonal subspace while largely preserving the dominant semantic direction. This structural flattening consistently emerges during the final decoding stage, forming a shared bottleneck across diverse generator architectures. Motivated by this signature, we propose LoRC, a framework that decouples semantic dominance to capture the collapsed residual geometry induced by the generative decoding bottleneck. Our method improves accuracy by an average of 7.0% across multiple benchmarks and achieves 97.0% accuracy on 39 unseen generators. These results demonstrate strong cross-model generalization and robustness, making LoRC a reliable approach for AIGI detection in complex real-world environments.

We frame Live Artifacts as a class of persistent generative media between static assets and interactive software. Unlike conventional generative outputs that collapse into static files, Live Artifacts retain their generative logic as a persistent media property, enabling continuous context-dependent regeneration. Time, location, or live data become part of their generative specifications, initiating coordinated updates across modalities (e.g., adapting text, visuals, and audio together) while preserving composition, semantics, identity, and cross-modal coherence. To facilitate experimentation with this medium, we present LiveCanvas, an authoring system that reconceptualizes visual layers as live generative specifications with explicit mutability and constrained dependencies. Creators orchestrate dynamic behaviors and manage generative persistence within a visual canvas rather than through programming, defining what remains stable, what can change, and how changes propagate. We evaluate Live Artifacts through a gallery of responsive examples and a qualitative study with six professionals, finding that LiveCanvas facilitates a shift from composing static outputs to crafting responsive generative artifacts while remaining aligned with familiar authoring practices.

The incorporation of Semantic Web technologies within scientific environments is becoming an increasingly popular Research Data Management (RDM) practice. While ontologies offer flexible, reusable and machine-readable vocabularies to describe domain-specific research data, Knowledge Graphs (KGs) facilitate the integration of heterogeneous data sources into an interoperable collection. Furthermore, KGs offer additional advantages, notably the use of expressive SPARQL queries, or the ability to define complex data validation rules with SHACL. This work describes the modelling of a relational RDM system with an ontology, and the subsequent construction of a KG based on it. Enabled by the highly interconnected nature of research data and experiment workflows present in the system, we not only show how we can easily and reliably build an efficient KG from such domain-specific RDM systems, but also how doing so enables more advanced use cases. This is demonstrated by the modelling of ideal counterparts for the experiment workflows logged in the KG, which are then used to programmatically generate SHACL shapes that fully validate the conformance of the latter. By integrating this functionality within a UI, we allow researchers to plan, reuse, share, and track the progress of their daily experiments.

It has been conjectured that the operator norm of the Lyapunov operator induced by the Frobenius norm is always attained at a symmetric matrix. The conjecture is known to hold for all matrices of order at most five. We give an integer matrix of order seven for which the skew-symmetric restricted norm is strictly larger than the symmetric restricted norm. A rational separator and exact-arithmetic certificates establish the strict inequality without relying on floating-point computations. A direct-sum construction yields counterexamples in every order $n \geq 7$; the case $n = 6$ remains open.

Reliable traffic sign detection is a prerequisite for the global deployment of autonomous driving systems, where regulatory compliance and road safety depend on perceiving signs correctly across regions, ranges, and weather conditions. Despite recent progress, vision-based methods continue to face three fundamental limitations: poor cross-regional generalization due to high diversity across countries, degraded performance on small-object detection at long ranges (traffic signs occupy as little as $10{\times}10$ pixels at 200m), and fragile temporal tracking under the strongly non-linear perspective distortion that occurs as a vehicle approaches a sign. In this paper, we address the problem of robust, long-range, region-agnostic traffic sign perception by combining camera and Light Detection and Ranging (LiDAR) sensing. We present a multi-modal detection framework whose Intensity-Aware Deformable Fusion module aligns retro-reflective LiDAR cues with camera features, anchoring detection on geometric invariants rather than region-specific visual appearance. We further introduce a dual motion-model tracker that explicitly accounts for non-linear perspective transformations during vehicle approach, substantially improving temporal consistency over linear motion assumptions. Additionally, we develop a semantic attribute classification pipeline that estimates occlusion level, readability, sign embeddedness, and road relevance, providing actionable context to downstream planning. Extensive evaluation on our dataset, spanning 60+ countries and 2,500+ hours of driving data, shows that the proposed pipeline achieves an Object Miss Ratio (OMR) of 0.49% across 221,068 evaluation sequences, demonstrating globally generalizable traffic sign perception in commercial-grade autonomous driving systems.

A 4-bit quantized weight specifies a rounding cell rather than a single full-precision value. We introduce in-cell learning, a paradigm for writing new knowledge only within these cells, so that re-quantizing the served weights reproduces the released integer codes and scales exactly. CellFill implements this idea with bounded trainable positions inside frozen quantization cells and ships the update as a separate, subtractively revocable file. Across published NF4 and W4A16 releases of Qwen3 and Gemma from 1.7B to 32B parameters, CellFill writes 83-99% of a real-fact corpus while returning the stored code on every constrained weight. The injected facts generalize to paraphrases and composition, and answer 78-88% of selected PopQA questions that the released model misses. Sequential experiments show that rehearsal preserves earlier knowledge, whereas available room and new-task plasticity decline across updates. Consolidation re-quantizes the learned weights to produce a declared major version, restoring room at a measured capability cost. A six-task write-rehearse-consolidate cycle retains at least 92.8% of first learning in two 8B runs and records zero code violations over 6.9 billion constrained weights at every fold. These results define a version-management protocol in which minor updates preserve the released quantized artifact bitwise and major updates are explicit, measurable, and verifiable.

Mesh quality is very well studied and widely used to quantify a good mesh. In contrast, a systematic study of quality of meshfree point clouds is lacking. This gap makes it difficult to substantiate the common claim that generating a good-quality point cloud is easier than generating a good mesh. Various definitions of point cloud quality have been proposed, some of which have theoretical significance for proving convergence and error bounds, while others are used in computational studies. In this work, we compare and contrast existing point cloud quality metrics and introduce a few new ones. We conduct extensive numerical tests with a meshfree collocation method across a wide range of scenarios, including both elliptic and hyperbolic equations, 2D and 3D domains, and variations in parameters of the numerical method. Based on these tests, we assess which quality metrics best correlate with numerical error. Our findings reveal six metrics that consistently serve as reliable indicators of point cloud quality, while also demonstrating that several widely used metrics are poor predictors of accuracy.

Infrared small target detection is still challenging in remote sensing imagery, because the targets are extremely small, exhibit weak local contrast, and are often embedded in complex and highly variable backgrounds. In addition to these inherent difficulties, we observe that existing detectors often show unstable performance when the target scale changes or when the scene background varies. This scale- and scene-sensitive degradation indicates that current methods are insufficient in simultaneously preserving target structure during feature downsampling and maintaining discriminative local contrast under background shifts, which finally results in unbalanced detection performance across different conditions. To improve detection robustness, this paper proposes a Relative Degradation Aware Network (RDANet) for infrared small target detection. RDANet consists of two dedicated modules: Multi-Scale Anti-Alias Downsampling (MSAD) and Prototype-Guided Skip Memory (PGSM). MSAD introduces multi-scale anti-alias filtering together with pixel-fold aggregation to reduce aliasing effects during resolution reduction, so that target shape information can be better preserved while irrelevant background responses are suppressed. PGSM further enhances the skip features by retrieving patch-level prototypes from a shared memory and adaptively integrating them into the current representation, which helps maintain stable local contrast cues under diverse scene backgrounds. Experiments on three public benchmarks show that RDANet achieves the best performance on most evaluation metrics, while scale- and background-stratified evaluations indicate more stable behavior across target sizes and scene complexity. The code is available at https://github.com/BIT-RuiLiu/RDANet.

Predicting transition states (TS) in chemical reactions is crucial, as they provide insights into reaction mechanisms. Recent work on TS prediction have focused on flow matching supervised on straight linear paths that do not align with actual reaction trajectories. We propose a novel flow matching-based framework ReCurveflow that learns to predict TS geometries supervised on continuously curved reference paths interpolated from a full NEB-derived band of molecular geometries. We also introduce off-path correction, which grants ReCurveflow with the ability to produce corrective velocity fields when engaged off-path geometry states during inference rollout, leading to better resistance against exposure bias and accuracy in TS prediction. Across three data splits and six evaluation metrics, ReCurveflow achieves the best result on the majority of split-metric combinations against seven baselines. Qualitative analyses further show that ReCurveflow generates reaction trajectories with energy profiles that closely track the reference NEB path, provides initializations that ease the NEB optimization bottleneck, and exhibits the intended corrective behavior in its learned velocity fields. The ReCurveflow codebase is publicly available at https://github.com/dmis-lab/ReCurveflow.

Document processing pipelines traditionally cascade optical character recognition (OCR) engines with downstream models for structured information extraction, leading to multi-stage error propagation. We fine-tune SmolDocling, a compact 256M-parameter vision-language model (VLM), to perform end-to-end key-value extraction directly from document images, jointly solving identification, localization, and association in a single pass without OCR preprocessing. We extend DocTags with specialized key, value, region, and link tags, enabling many-to-many relationships in a unified output sequence. To address data limitations, we design an augmentation pipeline combining synthetic form filling and graph-based crops that preserve complete key-value subgraphs. We further introduce a layout-aware evaluation framework extending text matching with spatial bounding box verification. On FUNSD, XFUND, and a large-scale private dataset, our model outperforms larger zero-shot VLM baselines under layout-aware evaluation, while being 27 times smaller than Qwen2.5-VL (7B) and over 5 times faster at inference. The model weights will be released publicly after publication.

Accurate electricity price forecasting is critical for smart grid stability, yet the heavy reliance on historical price lags in modern predictive models often obscures the fundamental physical drivers of market volatility. This paper proposes a regime-sensitive, explainable artificial intelligence (XAI) framework to unmask the hidden roles of load, climate, and calendar variables in the ERCOT Day-Ahead Market (2014-2024). Utilizing a Histogram-based Gradient Boosting Regressor (HGBR), we introduce a Suppression Ratio to quantify how price lags overshadow physical parameters. Our analysis reveals that while lags provide short-term memory during normal conditions, they fail to capture extreme dynamics; notably, their inclusion increased forecasting error (MAE/RMSE) during price spikes. By isolating these effects, the study revealed that price spikes are not merely the result of continuously rising temperatures, but rather emerge from the system reacting sharply within a specific temperature range. These findings underscore the importance of moving beyond lag-based models to uncover the true drivers of extreme price events.

Artificial Intelligence (AI) offers significant potential for future aviation systems; however, its integration into safety-critical applications requires compliance with the aviation sector's stringent safety standards. For AI and Machine Learning (ML)-based systems, the European Union Aviation Safety Agency (EASA) emphasizes the need to demonstrate the representativeness and completeness of the Operational Design Domain (ODD) and the associated data distributions used during development and verification. Despite this requirement, a structured engineering process for defining target distributions and evaluating representativeness within ODDs remains largely unexplored. This work presents a method for representativeness assessment of AI/ML constituent ODDs in the context of aviation safety assurance. Starting from the methodical identification of suitable target distributions, a process flow is proposed that guides developers from ODD definition and parameter distribution modeling to the quantitative assessment and interpretation of coverage results with respect to EASA's learning assurance objectives. As quantitative measures, the chi-squared goodness-of-fit test is examined and found unsuitable for the large data sets arising in this setting, leading to the adoption of the Kullback–Leibler divergence and Cramér's $V$ for the representativeness assessment. The method is demonstrated using the example of AI-based airborne collision avoidance, employing experimental data from previous Horizontal Collision Avoidance System (HCAS) and Vertical Collision Avoidance System (VCAS) simulations. The results illustrate how statistical distribution comparison methods can support the assessment of representativeness for safety-critical AI applications and contribute toward a systematic Safety-by-Design AI engineering process aligned with emerging EASA guidance.

XR systems expose timing quantities, such as motion-to-photon latency, frametime, or component-level runtime timings, that can be observed repeatedly as temporally ordered timing traces. Conventional reporting with means, standard deviations, percentiles, or histograms is useful, but it discards temporal ordering. We propose a general structure-aware methodology for analyzing and reporting XR timing traces. Each trace is represented by a compact, interpretable time-series signature, and collections of signatures can be visualized and compared statistically. We evaluate the method using engine-level application frametime traces from a large-scale in-the-wild VR dataset and compare timing signatures across HMD-labelled groups. Across multiple sampling and content-control conditions, structure-aware signatures reveal substantially stronger systematic multivariate differences between HMD-labelled groups than distribution-only summaries. A within-trace temporal-order shuffle control reduces this separation, particularly under content matching, providing direct evidence that original temporal ordering contributes information to the timing signatures. The strongest individual feature contributions vary across sampling and content-control conditions, indicating that no single timing characteristic dominates across analysis settings. Although demonstrated on application frametime, the representation operates on timing traces and therefore provides a basis for future application to other XR timing quantities, including instrumented motion-to-photon measurements.

Transportation networks, in particular multi-class transportation networks (i.e., networks with mixed vehicle types), are complex systems that are challenging to control. Recently, Deep Reinforcement Learning (DRL), which learns control policies from interactions with the environment, and Model Predictive Control (MPC), which uses a system model to optimize control inputs, have been increasingly utilized for transportation network control. However, nonlinear system dynamics and high-dimensional state spaces in large-scale networks limit DRL's learning capacity under time-constrained training and increase MPC's computation time, hindering real-time implementation with limited computational resources. Moreover, MPC depends on an accurate network model, which is often unavailable for complex systems such as multi-class transportation networks. This paper proposes a novel DRL-MPC framework for multi-class transportation networks that divides control authority between DRL and MPC, combining DRL's fast online computation and model independence with MPC's built-in optimization and constraint-handling capabilities. In the hierarchical framework, MPC operates at the higher level and determines low-frequency control inputs whose slower update rate accommodates its high computation time, while DRL operates at the lower level and determines high-frequency control inputs using its fast online deployment. The framework is evaluated on a multi-class freeway network against a hierarchical MPC controller and a hybrid state-feedback-MPC controller, including scenarios with model mismatch and noisy traffic demands. Results show that the proposed framework outperforms the hybrid state-feedback-MPC controller, substantially reduces online computation time compared with the hierarchical MPC controller, and provides more effective constraint enforcement under model mismatch.

We develop a unified information-theoretic framework for lower bounding minimax quantiles. The starting point is a loss-adapted Neyman–Pearson metaconverse that bounds the minimax success probability at every loss threshold and confidence level. The bound separates the small-ball behaviour of the prior under the loss from the statistical distinguishability of the observation model, and is optimised over an auxiliary output distribution. Different relaxations of this Neyman–Pearson bound yield converses based on \(f\)-informativity, Sibson mutual information \(I_\alpha\), Maximal Leakage, and Amemiya norms. Classical Fano and Le Cam lower bounds are recovered as special cases. The framework also clarifies why different information measures are suited to different recovery criteria. Maximal Leakage is exact for a class of symmetric exact-recovery problems. We use this identity to derive finite-sample bounds on the full minimax exact-recovery risk in the balanced Gaussian weighted stochastic block model, as well as two-sided finite-sample minimax-quantile bounds for low-rank matrix estimation under isotropic bounded-energy noise. For approximate Hamming recovery, we exhibit a heterogeneous binary model in which an optimised finite Sibson order yields a strong converse while the Maximal Leakage specialisation is trivial. Finally, for one-coordinate Poisson localisation, a Bennett-type Young function used through its Amemiya norm recovers the exact success-probability scale, whereas classical Fano and fixed-power relaxations are strictly weaker. These results show that sharp converses for minimax quantiles require adapting the information measure to the recovery resolution, whether exact or approximate, and to the tail behaviour of the likelihood ratio.

Document-Level Relation Extraction (DocRE) relies heavily on costly manually annotated datasets, while large distant supervision resources such as DocRED distant remain underexploited due to noise. We show that a critical yet overlooked source of noise lies in structural inconsistencies within relational triples, including violations of ontology constraints and logical contradictions. We introduce an ontology-driven framework to quantify and enforce structural consistency in DocRE datasets. Our analysis reveals substantial structural noise in DocRED distant and demonstrates that such inconsistencies propagate to model predictions. Enforcing structural well-formedness during training significantly reduces logical contradictions and consistently improves generalization performance. These findings establish structural consistency as a missing axis of supervision in DocRE and highlight structural regularization as an effective strategy for leveraging distant data at scale.

Motivated by Bailey's four-step FFT algorithm (1989), we develop additive FFT techniques for polynomial evaluation over affine subspaces of binary extension fields. Our key insight is that the Taylor expansion with respect to vanishing polynomials of subspaces provides a structural counterpart to Bailey's matrix formulation. It decomposes an additive FFT (AFFT) into independent sub-AFFTs associated with the columns and rows of a matrix. We first present a general-basis AFFT that applies to any ordered basis and any split of the dimension, providing a unified baseline for measuring the gains from specialization. We then specialize the framework to the Cantor special basis and obtain two AFFT algorithms. The first supports an arbitrary decomposition of the AFFT dimension and exploits the Cantor special basis structure to perform the Taylor expansion stage without finite field multiplications. The second uses a decomposition that preserves the binomial form of the relevant subspace polynomials. It requires exactly $\frac{1}{2}n\log_2 n$ multiplications, together with a closed-form addition count determined by the binary representation of $m$. Our implementation results show that this algorithm is faster than the LCH AFFT over a Cantor special basis in 37 of the 42 configurations tested across two hardware platforms. This performance advantage stems from its fully recursive structure, which provides memory locality by design and avoids separate basis-conversion and evaluation stages. Finally, in a separate analysis, we formalize the notion of partial Cantor special bases and identify parameter regimes in which both the von zur Gathen-Gerhard algorithm and our general-basis AFFT require fewer additions and multiplications than the first Gao-Mateer algorithm.

Evaluation of long-context Large Language Models (LLMs) has advanced rapidly. However, most existing benchmarks are limited to the document level and focus mainly on high-resource languages, leaving many fine-grained challenges insufficiently evaluated. To address this gap, we present MGAL, the first multilingual, granularity- and position-aware long-context benchmark. MGAL is constructed from United Nations (UN) reports spanning 8K to 128K tokens across the six official UN languages. It covers four coherent levels of linguistic granularity (word, sentence, paragraph, and document) and further stratifies entries by their position within the document (begin, middle, and end), indexed at both the document and paragraph levels. This design enables systematic diagnosis of multilingual long-context comprehension across different granularities. Through extensive experiments and analyses, we find that: (1) LLMs perform well at word-level tasks but struggle with coarser-grained ones; and (2) Closed-source models retain a clear performance advantage in lower-resource languages. We further identify two new challenges: (1) Under local semantic crowding, where neighboring sentences share topics and entities, models tend to follow surface cues (e.g., connectives like “however” or repeated entities) rather than the discourse role of the sentence in surrounding context (e.g., background, outcome); and (2) A gap between fluency and consistency in generated outputs, where models produce text that reads smoothly but drifts from the source facts. In addition, we observe several patterns in line with prior studies, including reliance on nearby evidence and reuse of options under uncertainty.

This paper presents a reference-guided reinforcement learning framework to generate stand-up motion for a 29-DOF Unitree G1 humanoid on deformable soft ground, using a human demonstration recorded on hard ground. The terrain compliance is modelled using solref and solimp parameters from MuJoCo's rigid body soft-contact model. The rewards consists of (i) reference motion tracking through residual joint-position control and (ii) explicit recovery objectives such as pelvis height, torso uprightness, and the final posture. First, the policy is trained with the specified rewards considering hard ground. Next, the terrain stiffness is lowered by updating solref and the nominal surface penetration zone is expanded using solimp. Subsequent training enables the policy to adapt to the delayed support force generation due to significant surface penetration during contact-intensive phases while preserving the original demonstration pattern. The learned policy successfully completes the fallen-to-standing task in simulation, reaching the targeted pelvis height and uprightness, with a maximum contact penetration of approximately 40 mm during the process. The proposed method is demonstrated on two stand-up sequences and successfully achieves the final recovery objective on both hard and soft ground. Ablation studies show that reference tracking alone is insufficient for successful stand-up, and that explicit recovery rewards are essential.

Agentic engineering systems have shown strong performance on general-purpose benchmarks, yet their effectiveness in enterprise resource planning (ERP) domain-specific languages (DSLs) remains underexplored. We introduce BC-Bench, a benchmark designed to evaluate agentic engineering on real-world tasks in AL, the DSL for Microsoft Dynamics 365 Business Central. BC-Bench comprises 101 manually curated tasks extracted from two Microsoft-owned production repositories, reflecting authentic ERP development workflows. Adapting the SWE-Bench methodology, we address the unique constraints of the AL ecosystem—including limited public resources and complex environment provisioning. Beyond generating functional code, BC-Bench evaluates test generation and supports multimodal problem statements where visual context is commonly present. We evaluate multiple frontier models across two agent harnesses, utilizing multi-run metrics to account for nondeterminism. In the Bug Fixing category, under our evaluated settings, between-model differences in resolution rate are larger than differences between the two evaluated agent harnesses, and improvements reported on general-purpose benchmarks do not consistently transfer to AL. These results highlight the need for domain-specific evaluation.

Nearly every retrieval-augmented question-answering system in production ships with a hidden interpreter: on each query a language model re-derives the meaning of raw corpus text and then throws that work away. Cheaper models do not close the gap: per-token prices have fallen by orders of magnitude while inference spend has risen, because context volume grows faster than prices fall. This is the modern equivalent of the full-table scan, and the remedy is the one databases found fifty years ago: do the expensive work once, at write time, into a maintained structure that makes reads cheap. A corpus whose read pattern is known before it ever meets a user can and should be indexed too. We call the paradigm ingest-time semantic compilation (ISC): compile a corpus's meaning into a queryable substrate with two coupled layers - incrementally maintained embeddings, and atomic claims whose provenance is validated at compile time - and treat that substrate as a first-class database object with its own DDL, maintenance contract, migration contract, and cost model. Two existence proofs support it. Substrate upkeep scales with change rather than corpus size: incremental updates run 33.7x cheaper than reconstruction while tracking it to floating-point precision. And on a held-out sample of 500 broadcast-interview transcripts, compiled claims as the retrieval payload win all 32 budget-by-model cells: 85.2% correct from roughly 2.2k reader tokens against 72.5% from 16.3k for the best chunk configuration anywhere. The only baseline that keeps pace is a contextualized-chunk pipeline with hybrid retrieval and reranking, statistically indistinguishable from compiled claims at roughly twenty-one times the query-path tokens - and it reaches that parity, we argue, precisely because it has itself begun to compile. We close with the systems agenda this opens, from compilation planners to read planning.

Product catalogs underpin search, discovery, and recommendation in e-commerce, yet they are often attribute-sparse: the attributes shoppers and downstream systems rely on are either buried in unstructured content such as titles and images or missing from the catalog altogether. Manually enriching e-commerce catalogs is impractical given their scale and rapid growth. This paper introduces TRACE, a novel framework for automated catalog attribute enrichment using agentic Large Language Models (LLMs). A ScoutAgent triangulates multimodal evidence across merchant catalogs, syndicated feeds, and identity-matched web search to propose candidate attribute values with supporting evidence, while a JudgeAgent verifies the proposed value for each attribute value against its supporting evidence and decides whether to publish it or route it to human review. On an offline human evaluation dataset, TRACE's proposed attribute values were 98.2% accurate at 74.7% attribute coverage. Deployed in production on an industry-scale catalog, TRACE increased impression-weighted enrichment coverage across four business verticals by 90.4%. An online experiment subsequently showed that surfacing the enriched attributes on the product detail page increased checkout conversion by 0.48%.

This paper investigates the development of causal foundation models for bounding the effect of interventions and counterfactuals from observational data. We show that a canonical prior can be defined with full support over the space of structural causal models with discrete observables. With this canonical prior, we translate the problem of bounding counterfactuals into that of learning distributions over functions that map data (and possibly structural assumptions) to a causal query of interest. This extends the promising causal foundational modelling paradigm to the estimation of partially-identifiable causal effects, i.e., under unobserved confounding, where multiple values are equally compatible with the observed data and prior structural assumptions.

Recent advances in multimodal retrieval have improved the ability to retrieve information from visually rich documents such as PDFs and reports. However, existing benchmarks remain largely centered on English and provide limited coverage of Korean visual documents with complex structures. Furthermore, most existing Korean resources primarily evaluate single-page retrieval, failing to capture realistic scenarios that require evidence aggregation across multiple pages. To address these gaps, we introduce KoViDoRe, a benchmark for Korean visual document retrieval. The dataset is constructed from publicly available Korean documents with diverse layouts, including tables, figures, and multi-column structures. We develop a multi-stage data curation pipeline consisting of structured document parsing, synthetic query generation using both summary-based and context-based strategies, and relevance mapping with human verification. Using KoViDoRe, we evaluate a wide range of multimodal retrieval models and observe that current models struggle to effectively handle Korean visual document retrieval, particularly in settings involving structured content and diverse query types. Motivated by this finding, we further curate a large-scale training dataset, Ko-VDR Train Public, to support the development of retrieval models tailored to Korean visual documents. Together, KoViDoRe and Ko-VDR Train Public provide a unified benchmark and training resource for Korean visual document retrieval.

Watermarking diffusion language models (DLMs) requires mechanisms compatible with iterative parallel unmasking rather than autoregressive decoding. Existing sampling-based watermarking methods typically inject position-wise i.i.d. perturbations, which can be poorly aligned with DLM decoding dynamics and degrade generation quality. We propose SAC-Copula, a quality-preserving watermarking method for DLMs based on smooth, locally correlated Gumbel perturbation fields constructed via a Gaussian copula. We further develop a SAC-aware detector using covariance-aware filtering and native-sample calibration. Mechanism-level analysis shows that local correlation reduces latent perturbation roughness and better matches iterative refinement dynamics. Experiments on LLaDA show that SAC-Copula achieves a favorable quality-detectability trade-off compared with existing baselines. In particular, further evaluations on Dream-7B and additional datasets show that SAC-Copula substantially improves PPL tail stability over the i.i.d. Gumbel baseline, while maintaining strong low-FPR detectability and competitive overall generation quality. Additional token-edit stress tests further assess watermark robustness under controlled synchronization drift. Code is available at https://github.com/PunkyKnife/SAC-Copula.