arXiv:2604.10767v2 Announce Type: replace
Abstract: Detecting vulnerabilities in source code remains critical yet challenging, as conventional static analysis tools construct inaccurate program representations, while existing LLM-based approaches often miss essential vulnerability context and lack grounded reasoning. In this paper, we introduce VulWeaver, a novel LLM-based approach that weaves broken program semantics into accurate representations and extracts holistic vulnerability context for grounded vulnerability detection. VulWeaver first constructs an enhanced unified dependency graph (UDG) by integrating deterministic rules with LLM-based semantic inference to address static analysis inaccuracies. It then extracts holistic vulnerability context by combining explicit contexts from program slicing with implicit contexts, including usage, definition, and declaration information. Finally, VulWeaver employs meta-prompting with vulnerability type specific expert guidelines to steer LLMs through systematic reasoning, aggregated via majority voting for robustness. Extensive experiments on PrimeVul4J dataset show that VulWeaver achieves a precision of 0.82, recall of 0.71, and F1-score of 0.76, outperforming state-of-the-art learning-based, LLM-based, and agent-based baselines by 25%, 17%, and 21% in F1-score, respectively. Notably, VulWeaver attains a VP-S score of 0.58, 164% higher than the best baseline, confirming its strong discriminative power in distinguishing vulnerable code from patched counterparts. VulWeaver also demonstrates cross-language generalizability on the C/C++ PrimeVul dataset with minimal adaptation, achieving an F1-score of 0.78. For practical usefulness, VulWeaver detected 26 true vulnerabilities across 9 real-world Java projects, with 15 confirmed by developers and 5 CVE identifiers assigned. In industrial deployment, VulWeaver identified 40 confirmed vulnerabilities in an internal repository.
Science Journals
arXiv:2604.11996v4 Announce Type: replace
Abstract: Should we trust Large Language Models (LLMs) with high accuracy? LLMs achieve high accuracy on reasoning benchmarks, but correctness alone does not reveal the quality of the reasoning used to produce it. This highlights a fundamental limitation of outcome-based evaluation: models may arrive at correct answers through flawed reasoning, and models with substantially different reasoning capabilities can nevertheless exhibit similar benchmark accuracy, for example due to memorization or over-optimization. In this paper, we ask: given existing benchmarks, can we move beyond outcome-based evaluation to assess the quality of reasoning itself? We seek metrics that (1) differentiate models with similar accuracy and (2) are robust to variations in input prompts and generation configurations. To this end, we propose a reasoning score that evaluates reasoning traces along dimensions such as faithfulness, coherence, utility, and factuality. A remaining question is how to aggregate this score across multiple sampled traces. Naively averaging them is undesirable, particularly in long-horizon settings, where the number of possible trajectories grows rapidly, and low-confidence correct traces are more likely to be coincidental. To address this, we introduce the Filtered Reasoning Score (FRS), which computes reasoning quality using only the top-K% most confident traces. Evaluating with FRS, models that are indistinguishable under standard accuracy exhibit significant differences in reasoning quality. Moreover, models with higher FRS on one benchmark tend to perform better on other reasoning benchmarks, in both accuracy and reasoning quality. Together, these findings suggest that FRS complements accuracy by capturing a model's transferable reasoning capabilities. We open source our evaluation codebase: https://github.com/Manas2006/benchmark_reproducibility.
arXiv:2604.14772v3 Announce Type: replace
Abstract: We study the finite-size Kuramoto model of all-to-all coupled phase oscillators with heterogeneous natural frequencies and characterize the minimal coupling strength required for the existence of a fully phase-locked equilibrium (in a co-rotating frame). To remove the degeneracy due to uniform phase shifts, we move to a reduced co-rotating frame and assess stability through the Jacobian of the reduced system: a fully phase-locked state is stable when this Jacobian is negative definite. This defines a stability region in the phase space. The Kuramoto vector field maps this region to a convex set in frequency space, so a fully-locked state at coupling $K$ exists exactly when the rescaled frequency vector $\hat{\mathbf{\omega}}/K$ lies inside that convex image. The critical coupling $K_{\ell}$ is defined as the smallest coupling strength for which a fully phase-locked equilibrium exists; geometrically, it corresponds to the first intersection of the ray $t\hat{\mathbf{\omega}}$ with the boundary of this convex set. Building on this convex-geometric structure, we construct an explicit polytope from analytically computable boundary points of the stability region, providing a closed-form upper bound $K_b \ge K_{\ell}$. The bound is exact for frequencies aligned with polytope vertices and offers a fully explicit outer approximation for general frequency vectors. While not uniformly sharp in a quantitative sense, this construction exposes the underlying geometry of stable fully phase-locking solutions. These results provide a practical use the convex-geometric structure underlying stable fully-locked states in the Kuramoto model.
arXiv:2604.16621v3 Announce Type: replace
Abstract: Shell structures are pivotal in the fields of architecture and engineering, due to their aesthetic appeal and structural efficiency. Recently, 3D concrete printing has reignited the interest in these structures. But, as printed concrete cannot be reinforced with steel, structures built in this way must be designed to withstand primarily pure compression: they must be funicular shells. Nevertheless, a fundamental challenge remains unsolved since Robert Hooke's discovered the catenary arch in 1675: it is not known whether the concept of a funicular polygon can be generalised to three-dimensional structures. Generative Adversarial Networks (GANs), have shown remarkable success in generating realistic data samples matching the distribution of the training data and have been shown to produce highly convincing synthetic images. This work proposes a physics-informed generative adversarial framework for the design of funicular shell structures. The approach employs a modified Deep Convolutional Generative Adversarial architecture physically guided by an auxiliary discriminator to generate realistic and structurally efficient shell geometries. Specifically, the model is constrained by the membrane factor to penalize geometries dominated by bending. An additional discriminator is also employed allowing the model to deal with more complex structures. Results show that the developed model is stable and capable of generating physically optimal, previously unseen, funicular shells with smooth forms and high membrane factor distributions.
arXiv:2604.16855v2 Announce Type: replace
Abstract: Camouflaged object detection (COD) segments objects that intentionally blend with the background, so predictions depend on subtle texture and boundary cues. COD is often needed under tight on-device memory and latency budgets, making low-bit inference highly desirable. However, COD is unusually hard to quantize aggressively. We study post-training W4A4 quantization of Transformer-based COD and find a task-specific cliff: heavy-tailed background tokens dominate a shared activation range, inflating the step size and pushing weak-but-structured boundary cues into the zero bin. This exposes a token-local bottleneck -- remove cross-token range domination and bound the zero-bin mass under 4-bit activations. To address this, we introduce COD-TDQ, a COD-aware Token-group Dual-constraint activation Quantization method. COD-TDQ addresses this token-local bottleneck with two coupled steps: Direct-Sum Token-Group (DSTG) assigns token-group scales to suppress cross-token range domination, and Dual-Constraint Range Projection (DCRP) projects each token-group clip range to keep the step-to-dispersion ratio and the zero-bin mass bounded. Across four COD benchmarks and two baseline models (CFRN and ESCNet), COD-TDQ consistently achieves an $S_\alpha$ score more than 0.12 higher than that of the state-of-the-art quantization method without retraining. The code is available at https://github.com/MCG-NKU/nku-model-compre.
arXiv:2607.13653v1 Announce Type: new
Abstract: Real-world deployment of embodied agents requires active exploration, visual grounding, and interactive intent disambiguation. However, existing frameworks often rely on privileged simulator states or assume complete instructions, bypassing realistic deployment challenges. To bridge this gap, we present REAL, an agentic framework for open-world mobile manipulation. REAL establishes sim-to-real-consistent environment APIs without oracle perception and integrates a simulated user to enable human-in-the-loop interaction. Within this environment, we design diverse task compositions to drive data collection, supervised fine-tuning, and online reinforcement learning, systematically optimizing agent performance. To comprehensively evaluate this approach, we introduce REAL-Bench, a benchmark spanning 241 tasks across active exploration, visual distraction, articulated manipulation, and interactive disambiguation.
Experimental results demonstrate that our trained agent outperforms leading commercial closed-source VLMs on interactive tasks with a 56.9% success rate. Further empirical analysis reveals that our hierarchical training pipeline successfully aligns the model's tool-use capabilities while maintaining robust open-vocabulary reasoning under extended exploration horizons. Finally, we deploy and evaluate our framework on a physical dual-arm mobile robot, where it achieves a 78.3% end-to-end success rate over 60 real-world episodes. These physical trials demonstrate robust zero-shot transferability to unseen household scenarios, validating that our sim-to-real-consistent design successfully bridges the reality gap for long-horizon mobile manipulation. Code is available at https://github.com/InternRobotics/REAL.
arXiv:2604.21400v4 Announce Type: replace
Abstract: 3D Gaussian Splatting (3DGS) has revolutionized neural rendering, yet existing methods remain predominantly research prototypes ill-suited for production-level deployment. We identify a critical "Industry-Academia Gap" hindering real-world application: unpredictable resource consumption from heuristic Gaussian growth, the "sparsity shield" of current benchmarks that rewards hallucination over physical fidelity, and severe multi-sensor data pollution. To bridge this gap, we propose YOGO (You Only Gaussian Once), a system-level framework that reformulates the stochastic growth process into a deterministic, budget-aware equilibrium. YOGO integrates a novel budget controller for hardware-constrained resource allocation and an availability-registration protocol for robust multi-sensor fusion. To push the boundaries of reconstruction fidelity, we introduce Immersion v1.0, the first ultra-dense indoor dataset specifically designed to break the "sparsity shield." By providing saturated viewpoint coverage, Immersion v1.0 forces algorithms to focus on extreme physical fidelity rather than viewpoint interpolation, and enables the community to focus on the upper limits of high-fidelity reconstruction. Extensive experiments demonstrate that YOGO achieves state-of-the-art visual quality while maintaining a strictly deterministic profile, establishing a new standard for production-grade 3DGS. To facilitate reproducibility, part scenes of Immersion v1.0 dataset and source code of YOGO has been publicly released. The project link is https://jjrcn.github.io/yogo-project-home/.
arXiv:2607.13609v1 Announce Type: new
Abstract: Recovering a latent potential from observed flow on a directed graph (a discrete Poisson problem with Dirichlet boundaries) is ill-posed, and the standard fix backfires: ridge regularization shrinks toward a gauge-meaningless origin, collapsing and reversing the recovered ordering ($+0.81\to-0.42$ rank correlation against a planted ground truth). The gauge-invariant graph Dirichlet energy removes the hazard and delivers parameter-insensitivity: the estimate is stable across four orders of magnitude in $\lambda$, whereas ridge inverts the ordering for every $\lambda>0$. We prove the reduced solve is SPD and preserves dynamic range exactly where ridge collapses it, and localize absorbing boundaries from flow alone via a Poisson residual. The $H^1$ seminorm is classical; what is new is the gauge diagnosis, the parameter-insensitivity it buys, and an ablation showing the result is robust to the extraction method. On three public clickstream corpora the gauge-invariant estimate retains $28$--$41\%$ of the interior dynamic range while ridge collapses to as little as $0.2\%$. The same gauge invariance carries into graph neural networks -- neutralizing the constant mode per layer prevents the oversmoothing that collapses a deep directed GCN -- linking this classical inverse problem to a central question in graph learning.
arXiv:2604.21688v2 Announce Type: replace
Abstract: The IC3 algorithm represents the state-of-the-art (SOTA) hardware model checking technique, owing to its robust performance and scalability. A significant body of research has focused on enhancing the solving efficiency of the IC3 algorithm, with particular attention to the inductive generalization process: a critical phase wherein the algorithm seeks to generalize a counterexample to inductiveness (CTI), which typically is a state leading to a bad state, into a broader set of states. This inductive generalization is a primary source of clauses in IC3 and thus plays a pivotal role in determining the overall effectiveness of the algorithm.
Despite its importance, existing approaches often rely on fixed inductive generalization strategies, overlooking the dynamic and context-sensitive nature of the verification environment in which spurious counterexamples arise. This rigidity can limit the quality of generated clauses and, consequently, the performance of IC3.
To address this limitation, we propose a lightweight machine-learning-based framework that dynamically selects appropriate inductive generalization strategies in response to the evolving verification context. Specifically, we employ a multi-armed bandit (MAB) algorithm to adaptively choose inductive generalization strategies based on real-time feedback from the verification process. The agent is updated by evaluating the quality of generalization outcomes, thereby refining its strategy selection over time.
Empirical evaluation on a benchmark suite comprising 914 instances, primarily drawn from the latest HWMCC collection, demonstrates the efficacy of our approach. When implemented on the state-of-the-art model checker rIC3, our method solves 26 to 50 more cases than the baselines and improves the PAR-2 score by 194.72 to 389.29.
arXiv:2604.23620v2 Announce Type: replace
Abstract: We present Move-Then-Operate, a Vision language action framework that explicitly decouples robotic manipulation into two distinct behavioral phases: coarse relocation (move) and contact-critical interaction (operate). Unlike monolithic policies that conflate these heterogeneous regimes, our architecture employs a dual-expert policy routed by a learnable phase selector, introducing a structural inductive bias that isolates phase-specific dynamics. Phase labels are automatically generated via an MLLM-based pipeline conditioned on lightweight contextual cues such as end-effector velocity and subtask decomposition to ensure alignment with human motor patterns. Evaluated on the RoboTwin2 benchmark, our method achieves an average success rate of $68.9\%$, outperforming the monolithic $\pi_0$ baseline by $24\%$. It matches or exceeds models trained on $10\times$ more data and reaches peak performance in $40\%$ fewer training steps, demonstrating that architectural disentanglement of move and operate phases is a highly effective and efficient strategy for mastering high-precision manipulation.
arXiv:2604.26059v3 Announce Type: replace
Abstract: Quantum Bayesian networks provide a mathematical formalism to describe causal relations, to analyse correlations, and to predict the probabilities of measurement outcomes, in systems involving both classical and quantum data. They generalize Pearl's Bayesian networks -- prominent graphical models for classical probabilistic reasoning and inference.
The goal of this paper is to bring compositional principles and a typing discipline into this setting. A key feature of our compositional semantics is that when all causes are classical, it coincides with the standard factor-based semantics of Bayesian networks, while in the purely quantum case it reduces to tensor networks. We then propose a typed formalism based on linear logic proof-nets, where types ensure well-behaved composition of systems, and which we prove sound and complete with respect to quantum Bayesian networks.
arXiv:2604.27232v3 Announce Type: replace
Abstract: Models of sign language have historically lagged behind those for spoken language (text and speech). Recent work has greatly improved their performance on tasks like sign language translation and isolated sign recognition. However, it remains unclear to what extent existing models capture various linguistic phenomena of sign language, and how well they use cues from the multiple articulators used in sign language (hands, upper body, face). We introduce a new benchmark dataset for American Sign Language, ASL Minimal Translation Pairs (ASL-MTP), divided into multiple types of sign language phenomena and corresponding minimal pairs of translations, for performing such linguistic analyses. As a case study, we use ASL-MTP to analyze a state-of-the-art ASL-to-English translation model. We conduct a targeted analysis of the model by ablating various input cues during training and inference and evaluating on the phenomena in ASL-MTP. Our results show that, while the model performs above chance level on most of the phenomena, it relies strongly on manual cues while often missing crucial non-manual cues.
arXiv:2604.27792v5 Announce Type: replace
Abstract: Vision-Language-Action (VLA) models generalize semantically well but often lack fine-grained modeling of world dynamics. We present Motubrain, a unified World Action Model that jointly models video and action under a UniDiffuser formulation with a three-stream Mixture-of-Transformers architecture. A single model supports policy learning, world modeling, video generation, inverse dynamics, and joint video-action prediction, while scaling to heterogeneous multimodal data such as video-only, task-agnostic, and cross-embodiment robot data. Building on Motus, Motubrain further introduces unified multiview modeling, an independent text stream for stronger language-action coupling, a shared cross-embodiment action representation, and an efficient post-training and deployment recipe for long-horizon real-world control. Our inference stack combines step reduction, compilation, FP8 quantization, DiT caching, V2A-style action-only inference, and real-time chunked closed-loop execution, achieving over 50x speedup over a naive baseline and up to 11 Hz inference. Experimentally, Motubrain achieves 95.8% and 96.1% average success on RoboTwin 2.0 under clean and randomized settings, respectively, attains the strongest reported EWMScore in our WorldArena comparison, and adapts to new humanoid embodiments with only 50--100 trajectories. These results show that unified world action models can scale in generality, predictive accuracy, and real-world deployability.
arXiv:2605.03268v2 Announce Type: replace
Abstract: Here we introduce Partially Observed Structural Causal Models (POSCMs) as an extension of structural causal models (SCMs) to settings where upstream contexts co-determine both the interaction structure and downstream mechanisms on observed variables. POSCMs thus provide a self-contained causal modeling framework for endogenous graphs, allowing for an intervention hierarchy spanning node- and edge-level contexts and endogenous variable interventions. To define edge interventions, we separate node mechanisms into edge-local transmission channels that can be modified without changing the source node or the rest of the target mechanism. We provide an identifiability theory that clarifies which intervention families would suffice to disentangle structure formation from mechanisms. We then empirically validate these theoretical results in two external simulators: a biophysically detailed virtual human retina and a gene-regulatory analogue. The experiments reproduce non-identifiability under latent context, expose structure-mechanism confounding under latent edges, and recover pathway-level input-output relationships under targeted interventions, consistent with our positive Markov kernel identifiability results. Together, POSCMs provide an intervention-oriented framework for causal systems in which contexts, graph structure, mechanisms, and measurements are jointly generated and only partially observed.
arXiv:2607.13654v1 Announce Type: new
Abstract: While 3D Gaussian Editing (3DGE) has seen substantial progress, text-driven 3D human garment editing remains largely underexplored. Existing 3DGE works typically follow a paradigm that applies 2D editing techniques to multi-view rendered images and updates 3D Gaussians based on the modified images. Extending such methods to 3D human garment editing suffers from low-fidelity outcomes, caused by introduced distortions and garment inconsistencies. A promising breakthrough opportunity arises from the SMPL eXpressive (SMPL-X) model that embodies rich prior information for virtual humans. Motivated by this insight, we propose a text-driven 3D human garment editor termed T3HG-Editor, which delivers high-fidelity and garment consistent results by leveraging geometry and joint priors embedded in SMPL-X. Specifically, T3HG-Editor contains three stages, namely obtainment of editable Gaussians, garment consistent editing, and Gaussian updating with overflow pruning. The obtainment of editable Gaussians begins with seeding Gaussians along SMPL-X normals to generate sufficient near surface Gaussians, followed by a 2D mask constraint that precisely localizes the target Gaussians to be edited. The garment consistent editing aggregates tokens corresponding to the same SMPL-X vertex across multiple views and propagates them to their original views, enforcing garment consistency without requiring additional training. Gaussian updating with overflow pruning employs a Signed Distance Function (SDF) defined on SMPL-X to construct a human distance field, which is then integrated with a 2D semantic mask to prune overflowing Gaussians, thus preventing contamination of non-target regions. Experiments on multiple subjects and diverse garment types demonstrate that T3HG-Editor outperforms state-of-the-art methods in both editing quality and garment consistency.
arXiv:2607.13655v1 Announce Type: new
Abstract: Explainable Reinforcement Learning (XRL) seeks to make Reinforcement Learning (RL) policies more transparent and interpretable, a key requirement in safety-critical and human-centric scenarios. However, it is mostly based on user studies, thus targeting the needs of a specific audience and lacking shared evaluation metrics. On the other hand, logic-based approaches within eXplainable Artificial Intelligence (XAI) provide compact, human-readable abstractions of decision-making. However, the systematic quantification of the explainability degree of logical representations remains an open problem. This work aims to advance the state of the art in XRL by introducing objective and planning-oriented metrics for policy explainability in RL settings. At the same time, it contributes to the field of logic for XAI by providing a principled way to quantify the explainability of logical rules, moving beyond common-sense assessments and simple propositional fragments. We employ Inductive Logic Programming (ILP) to extract symbolic representations of RL policies and define a novel set of explainability metrics, including activation rate, feature coverage, syntactic distance and semantic distance. These metrics quantify alignment between symbolic rules and agent behavior, the role of features in decision-making, and the evolution of policies during training and across agents in single and multi-agent RL. Experiments across different RL domains show that the proposed metrics highlight action-specific learning dynamics beyond global return, provide fine-grained insights into domain features beyond classical approaches for global feature importance estimation, and uncover coordination, specialization, and adaptation patterns in MARL. Moreover, they provide crucial insights for the transfer and generalization of action-specific policies.
arXiv:2607.13771v1 Announce Type: new
Abstract: We report a search for dark matter (DM) produced in association with a leptonically decaying \(Z\) boson at \(\sqrt{s}=13\) TeV using CMS Run 2015D open data corresponding to an integrated luminosity of \(2.32\,\mathrm{fb}^{-1}\) together with simplified-model Monte Carlo simulation. Events are selected in the mono-\(Z\rightarrow\ell^+\ell^-\) final state in both the \(\mu\mu\) and \(ee\) channels. Forty kinematic observables are extracted from MINIAOD and MINIAODSIM, cleaned with physics-motivated selections, and reduced to a 37-dimensional feature vector. Five Neural Spline Flows are trained independently to model Standard Model background and mediator-specific DM signal densities. The per-event test statistic is constructed from the log-likelihood ratio between the signal and background density estimates, providing sensitivity across the full kinematic phase space without requiring a hard upper \(\mathrm{MET}\) threshold. A simultaneous profile-likelihood fit combining the two channels yields observed (expected) 95\% confidence level upper limits on the signal-strength parameter of \(\mu<0.0177\) (\(0.0018\)) for the scalar mediator, \(\mu<0.0362\) (\(0.0039\)) for the vector mediator, and \(\mu<0.0498\) (\(0.0069\)) for the axial-vector mediator. The observed limits are weaker than expected because of a residual high-\(\mathrm{MET}\) background-modeling discrepancy rather than evidence for a DM signal. To our knowledge, this is the first application of Neural Spline Flow likelihood-ratio scoring to a mono-\(Z\) dark matter search using CMS Run 2015D open data simultaneously in the \(\mu\mu\) and \(ee\) channels.
arXiv:2605.07323v2 Announce Type: replace
Abstract: Discovering governing differential equations from observational data is a fundamental challenge in scientific machine learning. Existing symbolic regression approaches rely primarily on quantitative metrics; however, real-world differential equation modeling also requires incorporating domain knowledge to ensure physical plausibility. To address this gap, we propose DoLQ, a method for discovering ordinary differential equations with LLM-based qualitative and quantitative evaluation. DoLQ employs a multi-agent architecture: a Sampler Agent proposes dynamic system candidates, a Parameter Optimizer refines equations for accuracy, and a Scientist Agent leverages an LLM to conduct both qualitative and quantitative evaluations and synthesize their results to iteratively guide the search. Experiments on multi-dimensional ordinary differential equation benchmarks demonstrate that DoLQ achieves superior performance compared to existing methods, not only attaining higher success rates but also more accurately recovering the correct symbolic terms of ground truth equations. Our code is available at https://github.com/Bon99yun/DoLQ.
arXiv:2605.07914v2 Announce Type: replace
Abstract: Sharpness-aware and gradient-alignment methods have been shown to improve generalization, however each family of methods targets a single geometric property of the loss landscape, while ignoring the other. In this paper, we show that this omission is structurally unavoidable and that both flatness and gradient alignment should be considered in multi-distribution learning settings. Specifically, we derive an excess-risk decomposition that yields two additive leading-order terms: (i) an alignment term, controlled by the trace of $\bar{H}^{-1}\Sigma_g$ and (ii) a curvature term, controlled by $\bar{H}$, where $\bar{H}$ is the average Hessian and $\Sigma_g$ is the covariance of the gradient across distributions. Notably, $\bar{H}$ appears inverted in one and non-inverted in the other. We further show, via a counterexample, that neither quantity bounds the other in general, so no algorithm targeting only one term can guarantee low excess risk. Motivated by this decomposition, we propose SAGE (Spectral-Aware Gradient-Aligned Exploration) that targets both terms. The curvature component replaces SAM's gradient-scaled perturbation with the polar factor of each layer's gradient matrix, computed via Newton-Schulz iteration, so that the ascent step probes all directions with similar magnitude. On the other hand, the alignment component injects isotropic noise at the descent step, the magnitude of which scales with cross-distribution gradient disagreement. Experiments on five domain-generalization and two multi-task learning benchmarks show that the proposed method establishes a new state-of-the-art on DomainBed and acts as a general-purpose improvement to base MTL solvers, remaining competitive with, or even surpassing, state-of-the-art methods.
arXiv:2605.08583v2 Announce Type: replace
Abstract: Large language models are increasingly used in scientific writing, yet they can fabricate citation-shaped references that appear plausible but fail bibliographic verification. Existing detectors often reduce verification to binary found/not-found decisions and rely on brittle parsing or incomplete retrieval, offering little field-level signal to auditors. We reframe citation hallucination detection as taxonomy-aligned field-level adjudication and introduce a 12-code taxonomy spanning Real, Potential, and Hallucinated citations. Based on this taxonomy, we build CiteTracer, a cascading multi-agent detector that extracts structured citations from PDF and BibTeX, retrieves evidence through cache lookup, URL fetch, scholar connectors, and web search, applies deterministic field matching, and routes ambiguous cases to class-specialist judgers. We release a benchmark of 2,450 synthetic citations built from real seeds with controlled LLM mutations, paired with 957 real-world fabricated citations drawn from ICLR 2026 and an anonymous conference desk-rejected submissions. CiteTracer reaches 97.1% accuracy on the synthetic benchmark, with class-level F1 scores of 97.0, 95.8, and 98.5 for Real, Potential, and Hallucinated, respectively, and detects 97.1% of fabrications on the real-world set without abstaining. Code: https://github.com/aaFrostnova/CiteTracer.
arXiv:2605.10444v2 Announce Type: replace
Abstract: Understanding the molecular mechanisms by which denaturants modulate protein structure remains a central challenge in protein biophysics. In this study, molecular dynamics (MD) simulations were employed to investigate the effects of urea on the structural stability of bovine serum albumin (BSA) in its F isoform at pH 3.7 across a broad range of urea concentrations, from pure aqueous solution (0 M) to a fully urea-solvated environment. The simulations reveal a concentration-dependent remodeling of the protein hydration shell. At low urea concentrations, backbone-water hydrogen bonds decrease by approximately 40 %t, accompanied by an approximately 45 % increase in protein-urea hydrogen bonds between 1 and 7 M urea, consistent with a competitive solvation process in which urea progressively replaces water molecules at the protein surface. As urea concentration increases, urea-urea self-association becomes increasingly significant, reducing the number of direct protein-urea contacts; concurrently, the remaining water molecules form protein-water hydrogen bonds more efficiently on a per-water-molecule basis, without implying a net increase in the absolute number of hydration water molecules. Despite these pronounced solvent rearrangements, the secondary structure of BSA remains largely preserved throughout the simulations. In contrast, local structural organization and global conformational features, particularly within Domain III, exhibit increased solvent exposure and enhanced conformational flexibility.
arXiv:2607.13661v1 Announce Type: new
Abstract: Contrastive Language-Image Pre-training (CLIP) has been shown to have limitations in its fine-grained dense feature representation, due to its pre-training focusing on matching the whole image to a text description. Considering the large data and computational burden in pre-training a vision-language model from scratch, a series of works aim to enhance the fine-grained ability of CLIP through a fine-tuning scheme. However, existing works suffer from a variety of limitations: additional region annotations are usually required, which limits the semantic diversity due to the predefined categories and leads to a large effort to process the training data; and they usually sacrifice CLIP's original ability for global visual representation. To bypass these limitations, we propose SFF-CLIP (Self-annotated Fine-grained Fine-tuning for CLIP), which only uses image-text pairs as input to boost the fine-grained representation ability in the CLIP fine-tuning, while maintaining the global visual-semantic consistency. Concretely, a run-time region-phrase alignment scheme is designed, which obtains concept phrases from the input sentence, and aligns them with corresponding extracted region-based features using text-specific heat maps. Extensive experiments demonstrate that SFF-CLIP leads to significant performance improvements on fine-grained dense feature representation, as well as maintaining the performance of the original CLIP on image-level tasks. Code will be released later.
arXiv:2607.13077v1 Announce Type: new
Abstract: Large language models (LLMs) often produce homogeneous outputs, raising concerns that AI coding assistants may lead to convergence in the software artifacts that developers create. Whether this occurs in practice is unclear because developers interactively prompt, evaluate, modify, and reject model outputs, and because outputs vary with prompt and repository context. I examine code homogenization using Kaggle contest submissions from 2019 to mid-2026. I first document widespread convergence toward the random seed value 42, consistent with LLMs reinforcing a longstanding convention in programming culture. I then study homogenization more broadly, at two levels of aggregation and abstraction. At the submission level, I measure the average pairwise similarity of submissions within contests. At the contest level, I measure the conceptual span of submitted code, motivating distinct measures for each: TF-IDF representations, which capture surface syntax, and Voyage 3 code embeddings, which capture code intent and semantics. The results demonstrate substantial syntactic homogenization at both the individual and collective levels: individual submissions have become more alike in literal syntax and code structure, while the latent dimensionality of syntactic variation has narrowed. In contrast, I find little evidence of semantic homogenization, individually and collectively. Average semantic distance remains essentially flat, and the contest-level latent dimensional span of semantic approaches remains stable, with evidence suggesting it has even expanded modestly. These findings suggest that AI coding assistants are certainly standardizing implementation details, yet they have not yet produced evidence of homogenization in the approaches and problem-solving strategies coders employ.
arXiv:2607.13344v1 Announce Type: new
Abstract: Constraint Answer Set Programming (CASP) is a hybrid reasoning paradigm that combines Answer Set Programming (ASP) with Constraint Processing and Satisfiability Modulo Theories (SMT), enabling powerful declarative encodings of complex combinatorial search problems. This paper presents the design and implementation of EZSMTV3, an extensible SMT-based CASP framework that advances the translational approach to CASP solving. Building upon the foundation of the EZSMT+ system, EZSMTV3 introduces a more expressive input language, supports optimization via weak constraints, and offers foundations for streamlined integration of new constraint types. Rather than implementing custom search procedures, EZSMTV3 leverages state-of-the-art SMT solvers, such as CVC5, YICES, and Z3 to perform reasoning. The paper provides benchmarking results comparing EZSMTV3 with its CASP peers such as CLINGCON, CLINGO[DL], and CLINGO[LP], while showcasing its ability to handle mixed-domain constraints involving both integers and reals. The system provides a robust platform for future extensions and theoretical exploration within the CASP domain.
arXiv:2607.13078v1 Announce Type: new
Abstract: LLMs are now proposed for fraud detection, scam investigation, content moderation, and other trust-and-safety workflows. Much of the public literature still evaluates them as models, with less attention to their behavior as components in operational pipelines. This creates a practical evidence question: what would justify placing an LLM inside a live workflow with latency, cost, escalation, human-review, and adversarial-risk constraints?
We address this question through a fraud-first survey of deployment evidence. We code 49 operationally relevant sources on LLM use in fraud detection, investigation support, content moderation, and cross-cutting robustness (18 fraud, 14 moderation, 17 cross-cutting), supplemented by 15 contextual references that establish the survey boundaries. These sources include systems, benchmarks, frameworks, and deployment-relevant surveys, not 49 production deployments.
The main finding is an evidence imbalance. Fraud supplies the largest task-specific portion of the coded corpus. The moderation papers, however, include more explicit public evidence on latency, cost, governance, and fairness. Among the 18 fraud and investigation sources, none report clean per-decision latency, per-decision dollar cost, or calibration evidence; most report offline task performance, retrieval gains, or case-study accuracy instead.
The survey contributes a role-and-evidence organizing frame, FORTE, for locating LLMs as classifiers, retrieval interfaces, explanation generators, reviewer assistants, agents, feature extractors, or escalation components. It also contributes a minimum deployment-evidence checklist covering latency budget, cost per decision, decision threshold, explanation integrity, and adversarial pressure. The resulting agenda identifies studies needed to support deployment claims for LLM-based fraud and trust-and-safety work.