Academic Intelligence · Curated Daily

探索全球前沿学术脉络

AcademicHub 汇聚顶级期刊与预印本平台的实时文献。定制您的专属科研雷达,利用大语言模型自动生成交叉领域文献分析简报。

01.
arXiv (CS.CV) 2026-06-19

MMD-SLAM: Structure-Enhanced Multi-Meta Gaussian Distribution-Guided Visual SLAM

3D Gaussian Splatting (3DGS) has significantly boosted novel view synthesis and high-fidelity scene reconstruction, expanding the potential of 3DGS-based Visual Simultaneous Localization and Mapping (SLAM) methods. However, most existing systems fail to fully exploit the underlying structural information, which limits rendering quality and often leads to inconsistent maps. To address these limitations, we propose MMD-SLAM, a structure-enhanced Visual SLAM framework that leverages the Atlanta World (AW) assumption to guide a Multi-Meta Gaussian representation for photorealistic mapping. First, we introduce a point-line fusion strategy for pose optimization, where 3D line segments are incorporated to improve tracking robustness and provide additional constraints for mapping. Second, we design a Multi-Meta Gaussian representation with dominant directions, explicitly encoding structural priors from the AW hypothesis. Finally, we propose a Gaussian evolution strategy that adapts to scene geometry and incorporates structural cues into global optimization. Extensive experiments demonstrate that these innovations enable MMD-SLAM to achieve state-of-the-art performance in both tracking accuracy and mapping quality. e.g., our method achieves a 48.56% reduction in ATE RMSE on ScanNet and a 5.71% improvement in PSNR on Replica, compared with MonoGS.

02.
arXiv (CS.LG) 2026-06-16

Tail-Shape Estimation in LLM Evaluation Is Fragile: A Protocol for Diagnosing False Positives

作者:

arXiv:2606.16511v1 Announce Type: new Abstract: Recent work motivates moving large language model (LLM) evaluation from mean-based to tail-aware metrics, including conditional value-at-risk and tail-index estimates of reward-model error. We ask whether the canonical extreme-value-theory tail-index parameter, which isolates how heavy a tail is from how large the tail mass is, adds discriminative information beyond the mean and a standard tail-magnitude statistic in LLM evaluation. We pre-register a protocol covering admissibility, goodness-of-fit, threshold-stability, and effect-size requirements for any positive tail-shape claim. The protocol is the contribution of this paper; the empirical study below is a demonstration of what its gates catch. Applied to a standard LLM toxicity-evaluation setup under two structurally different scorer families, the protocol catches three distinct modes of false positives that a naive analysis would have published, and rejects the headline tail-shape claim on both scorers. We conclude that tail-shape estimation in the LLM toxicity-evaluation setups we examined is more fragile than the recent literature suggests, and recommend the protocol as a starting point for tail-index claims in similar setups.

03.
medRxiv (Medicine) 2026-06-19

"Us with them": Co-designing a caesarean section consent and debriefing intervention in West Cameroon

Background Women-centred maternity care is a rights issue that determines the use of services. Such care ensures responsiveness to womens needs which is enacted through shared decision-making, review and response. In the West Region of Cameroon, informed consent (IC) and Debriefing for caesarean section (c-section) have been shown to be suboptimal or absent. This paper describes the participatory design of a quality-improvement hospital-based intervention. Methods From February to May 2025, we conducted a co-design process with three groups of stakeholders: 59 post c-section women and community representatives, 78 frontline c-section providers, and 29 directors of public and private hospitals. We followed four phases: planning, conducting, evaluating, and reporting. The conduct phase comprised five all-day workshops with post c-section women and community representatives, followed by five all-day workshops with the c-section providers. Finally, we held an 11th workshop with the hospital directors to scrutinize suggested interventions, evaluate their feasibility, and establish a consensus on their components. We described the intervention using the TIDieR (Template for Intervention Description and Replication) checklist. We documented the co-design process, using open-ended narratives to delineate interventions, and carried out real-time synthesis on visual aids (whiteboards and flipcharts). Intervention feasibility was quantified using a structured ad hoc matrix, while insights on facilitators and barriers were captured through qualitative free-text entries. We coupled data collection with constant comparison and triangulation through contemporaneous field notes, photographic documentation, and thematic mapping of stakeholders perceptions and interactive dynamics. Results Participants perspectives on the co-design were positive, and their motivation were very high although less than 50% reported previous involvement in co-design processes. More than 80% of participants found rated the co-design process as either good or very good. The final intervention comprised four components: (i) an in-service training; (ii) a standard operating procedure including a harmonised consent form and debriefing checklist; (ii) systematic supportive supervision, monitoring & evaluation; and (iv) a routine clinical audit. Each group of stakeholders upheld specific dimensions of the consent and debrief intervention. Post c-section women and community members emphasized emotional support, written discharge advice after debriefing, and zero tolerance of suboptimal consent and debriefing practices. Frontline c-section providers insisted on robust documentation for medico-legal protection. Hospitals Directors emphasized capacity-building and cultural friendliness. All the groups supported womans autonomous decision making. The intervention feasibility was rated high or very high by hospital directors except for the financial, infrastructural and technical domains. Conclusion This co-design process yielded a context-specific, multi-component intervention that was well accepted and deemed feasible across stakeholders. It provides a methodological approach to strengthening informed consent and debriefing as core elements of women-centred, accountable maternity care, and warrants implementation.

04.
medRxiv (Medicine) 2026-06-15

VarEx: A Large Language Model Pipeline for Automated Extraction of Exposures, Outcomes, and Covariates from Epidemiologic Studies

Objective: Observational studies are essential for investigating risk factors for Alzheimer's disease and related dementias (ADRD), but inconsistent reporting and selection of covariates can contribute to residual confounding, omitted-variable bias, and reduced reproducibility. We developed and evaluated VAREX (Variable Extraction), a large language model (LLM)-based information extraction framework designed to automatically identify exposures, outcomes, and covariates from epidemiologic studies and populate structured evidence repositories. Materials and Methods: VAREX combines retrieval-augmented generation, biomedical language-model embeddings, semantic chunking, cross-encoder reranking, and prompt-engineered LLM workflows to extract epidemiologic variables from full-text biomedical articles. The framework was evaluated using a reference-standard corpus of observational studies examining blood pressure variability (BPV) and Alzheimer's disease-related dementias (ADRD), together with external validation datasets involving other exposure-outcome relationships. Extracted variables were compared with independently curated human reference standards using semantic matching and one-to-one assignment procedures. Covariates were additionally classified into ten epidemiologically relevant semantic categories. Results: In the primary BPV[->]ADRD corpus (10 studies), VAREX achieved a precision of 0.91, recall of 0.84, and F1-score of 0.87 for variable extraction. Covariate classification accuracy was 0.90, yielding a strict extraction-and-classification F1-score of 0.78. External validation datasets demonstrated comparable performance across diverse epidemiologic domains, with extraction F1-scores ranging from 0.73 to 0.85. Category-level performance was strongest for health behaviors (F1=0.96), sociodemographic variables (F1=0.90), and medication exposures (F1=0.89). Compared with published estimates of manual systematic-review effort, VAREX reduced processing time from approximately 61 minutes to 9 minutes per article, representing an 85.7% reduction in review time. Discussion: These findings demonstrate that LLM-based information extraction can accurately identify and classify epidemiologic variables across heterogeneous observational-study designs. Automated extraction enables scalable construction of structured repositories of exposures, outcomes, and covariates while substantially reducing the labor required for evidence synthesis and systematic reviews. Conclusion: VAREX provides an effective framework for automated extraction and classification of epidemiologic variables from the biomedical literature. By supporting large-scale evidence synthesis and structured knowledge resource development, VAREX may facilitate more rigorous observational research, improved confounder identification, and enhanced reproducibility in epidemiology.

05.
arXiv (quant-ph) 2026-06-16

Simulation of Non-Hermitian Hamiltonians with Bivariate Quantum Signal Processing

arXiv:2605.12450v2 Announce Type: replace Abstract: We achieve query-optimal quantum simulations of non-Hermitian Hamiltonians $H_{\mathrm{eff}} = H_R + iH_I$, where $H_R$ is Hermitian and $H_I \succeq 0$, using a bivariate extension of quantum signal processing (QSP) with non-commuting signal operators. The algorithm encodes the interaction-picture Dyson series as a polynomial on the bitorus, implemented through a structured multivariable QSP (M-QSP) circuit. A constant-ratio condition guarantees scalar angle-finding for M-QSP circuits with arbitrary non-commuting signal operators. A degree-preserving sum-of-squares spectral factorization permits scalar complementary polynomials in two variables. Angles are deterministically calculated in a classical precomputation step, running in $\mathcal{O}(d_R \cdot d_I)$ classical operations. Operator norms $\alpha_R\,,\beta_I$ contribute additively with query complexity $\mathcal{O}((\alpha_R + \beta_I)T + \log(1/\varepsilon)/\log\log(1/\varepsilon))$ matching an information-theoretic lower bound in the separate-oracle model, where $H_R$ and $H_I$ are accessed through independent block encodings. The postselection success probability is $e^{-2\beta_I T}\|e^{-iH_{\mathrm{eff}}T}|\psi_0\rangle\|^2\cdot (1 - \mathcal{O}(\varepsilon))$, decomposing into a state-dependent factor $\|e^{-iH_{\mathrm{eff}}T}|\psi_0\rangle\|^2$ from the intrinsic barrier and an $e^{-2\beta_I T}$ overhead from polynomial block-encoding.

06.
arXiv (CS.AI) 2026-06-16

MUZZLE: Adaptive Agentic Red-Teaming of Web Agents Against Indirect Prompt Injection Attacks

arXiv:2602.09222v2 Announce Type: replace-cross Abstract: Large language model (LLM) based web agents are increasingly deployed to automate complex online tasks by directly interacting with web sites and performing actions on users' behalf. While these agents offer powerful capabilities, their design exposes them to indirect prompt injection attacks embedded in untrusted web content, enabling adversaries to hijack agent behavior and violate user intent. Despite growing awareness of this threat, existing evaluations rely on fixed attack templates, manually selected injection surfaces, or narrowly scoped scenarios, limiting their ability to capture realistic, adaptive attacks encountered in practice. We present MUZZLE, an automated agentic framework for evaluating the security of web agents against indirect prompt injection attacks. MUZZLE utilizes the agent's trajectories to automatically identify high-salience injection surfaces, and adaptively generate context-aware malicious instructions that target violations of confidentiality, integrity, and availability. Unlike prior approaches, MUZZLE adapts its attack strategy based on the agent's observed execution trajectory and iteratively refines attacks using feedback from failed executions. We evaluate MUZZLE across diverse web applications, user tasks, and agent configurations, demonstrating its ability to automatically and adaptively assess the security of web agents with minimal human intervention. Our results show that MUZZLE effectively discovers 44 new attacks on 4 web applications with 10 adversarial objectives that violate confidentiality, availability, or privacy properties across different LLMs and agent scaffolds. MUZZLE also identifies novel attack strategies, including 3 cross-application prompt injection attacks and an agent-tailored phishing scenario.

07.
arXiv (CS.AI) 2026-06-11

GEAR-VLA: Learning Geometry-Aware Action Representations for Generalizable Robotic Manipulation

arXiv:2606.08530v2 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) models achieve strong benchmark performance but still struggle in real-world deployment with unseen objects, background shifts, and different robot embodiments. We argue that this stems from the lack of a unified geometry-aware manipulation representation, leaving existing VLAs vulnerable to low-level trajectory supervision, misaligned 3D features, and embodiment differences. To address this, we propose GEAR-VLA, a VLA framework for learning unified geometry-aware action representations for generalizable robotic manipulation. GEAR-VLA adopts coarse-to-fine action learning, where multi-source embodied pretraining equips the VLM with embodied reasoning and discrete action understanding before latent action tokens connect action semantics to a gradient-decoupled DiT continuous action expert. It further performs semantic-aligned 3D integration by aligning a trainable 3D spatial backbone with the VLA representation while freezing the original VLM-aligned visual pathway. To share this representation across robots, GEAR-VLA uses embodiment canonicalization, where embodiment-aware states and embodiment-invariant actions confine robot differences to the low-level interface. Extensive simulation and real-world experiments demonstrate strong generalization: GEAR-VLA achieves state-of-the-art performance on LIBERO, zero-shot LIBERO-Plus, and RoboTwin 2.0, reaches 85.9% success on AgileX and 81.0% on the pretraining-unseen LDT-01 embodiment, and obtains 90.1% success on a 6,360-trial universal grasping benchmark with 212 unseen objects. Code and models will be released at https://github.com/babynabeauty/GEAR-VLA.

08.
arXiv (CS.LG) 2026-06-12

LLM-ODDR: A Large Language Model Framework for Joint Order Dispatching and Driver Repositioning

arXiv:2505.22695v2 Announce Type: replace Abstract: Ride-hailing platforms face significant challenges in optimizing order dispatching and driver repositioning operations in dynamic urban environments. Traditional approaches based on combinatorial optimization, rule-based heuristics, and reinforcement learning often overlook driver income fairness, interpretability, and adaptability to real-world dynamics. To address these gaps, we propose LLM-ODDR, a novel framework leveraging Large Language Models (LLMs) for joint Order Dispatching and Driver Repositioning (ODDR) in ride-hailing services. LLM-ODDR framework comprises three key components: (1) Multi-objective-guided Order Value Refinement, which evaluates orders by considering multiple objectives to determine their overall value; (2) Fairness-aware Order Dispatching, which balances platform revenue with driver income fairness; and (3) Spatiotemporal Demand-Aware Driver Repositioning, which optimizes idle vehicle placement based on historical patterns and projected supply. We also develop JointDR-GPT, a fine-tuned model optimized for ODDR tasks with domain knowledge. Extensive experiments on real-world datasets from Manhattan taxi operations demonstrate that our framework significantly outperforms traditional methods in terms of effectiveness, adaptability to anomalous conditions, and decision interpretability. To our knowledge, this is the first exploration of LLMs as decision-making agents in ride-hailing ODDR tasks, establishing foundational insights for integrating advanced language models within intelligent transportation systems. While the current framework incurs higher computational costs than traditional methods, we show that parallel decomposition and model distillation can reduce latency to production-viable levels for deployment.

09.
arXiv (CS.AI) 2026-06-17

SoftMoE: Soft Differentiable Routing for Mixture-of-Experts in LLMs

arXiv:2606.17952v1 Announce Type: cross Abstract: Sparse Mixture-of-Experts (MoE) architectures enable scaling LLM parameters under a fixed inference budget by activating only a small subset of experts via top-$k$ routing. While this preserves causality and suits autoregressive language models, the discrete top-$k$ operator is not differentiable, forcing a fixed number of active experts per input and resulting in inefficient use of computation. We propose SoftMoE, which replaces discrete routing with a truncated soft top-$k$ LapSum relaxation, allowing gradient-based optimization of expert routing. We further parameterize the mean number of active experts per layer and impose a global budget constraint, enabling the model to learn how to allocate expert capacity across layers. SoftMoE remains fully compatible with autoregressive modeling and achieves performance comparable to or better than sparse MoE on language modeling and downstream tasks, while activating significantly fewer experts. Notably, the learned allocation is highly non-uniform, with later layers activating more experts. The source code is publicly available$^\dagger$.

10.
arXiv (CS.LG) 2026-06-17

Adaptable Method for Crystal Design across Diverse Constraints and Objectives with Pretrained Property Predictors

arXiv:2410.08562v5 Announce Type: replace-cross Abstract: Advanced crystal design can accelerate materials discovery across applications from photovoltaics to spintronics. Practical design must satisfy multiple properties and physical constraints, yet existing machine-learning-based approaches to such design often depend on large datasets, retraining, or task-specific generators. Here, we show that direct predictor-guided gradient optimization enables data-efficient, constraint-rich crystal design by combining off-the-shelf predictors with site-wise element masks, template initialization, and task-specific losses. In perovskites, it outperformed generative and Bayesian baselines under three targets – band gap, formation energy, and tolerance factor – and two hard constraints. DFT assessment further showed band-gap targeting competitive with a leading generative model despite using predictors trained on roughly one-tenth of the data. By flexibly combining pretrained predictors with application-oriented masks and custom losses, the same framework supported half-metal design. Such modularity could help researchers and engineers translate diverse application requirements directly into optimized candidate crystals with minimal computational cost.

11.
medRxiv (Medicine) 2026-06-22

Toward less intrusive pubertal assessment: longitudinal evaluation of tanner and non-tanner metrics in East African adolescents

Background: Accurate pubertal assessment is essential in pediatric endocrinology and adolescent health research. While Tanner staging remains the gold standard, its subjective nature and invasive genital examination limit feasibility and acceptability, especially in longitudinal studies and culturally sensitive settings. This study evaluated less intrusive pubertal assessment combinations that maintain discriminative accuracy. Methods: We conducted a longitudinal study among 200 uncircumcised, sexually naive males aged 15-17 years in Southwestern Uganda, with quarterly follow-up over three years. Clinicians assessed Tanner staging metrics (pubic hair, testicular volume, penile length, scrotal color), axillary hair, and serum testosterone. Markov transition models estimated Tanner stage progression. Ordinal logistic regression and area under the receiver operating characteristic curve (AUC) analyses quantified discriminative performance of individual and combined metrics. Results: At baseline, participants were distributed across Tanner stages II (6.0%), III (13.5%), IV (55.0%), and V (25.5%). Among individual metrics, pubic hair distribution best predicted overall Tanner stage (AUC=0.867), while penile length was least predictive (AUC=0.833). The full four-metric Tanner model achieved high discrimination (AUC=0.993). However, a less intrusive combination of pubic hair and scrotal color achieved comparable discrimination (AUC=0.942), improving to AUC=0.953 with axillary hair and age. Markov modeling demonstrated frequent bidirectional transitions between Tanner stages IV and V, reflecting variability in longitudinal staging. Conclusions: A minimally intrusive assessment combining pubic hair, scrotal color, axillary hair, and age reliably predicts pubertal stage, offering an acceptable alternative to traditional Tanner staging for research and surveillance contexts where genital manipulation is impractical or unethical.

12.
arXiv (CS.CL) 2026-06-16

SING: Synthetic Intention Graph for Scalable Active Tool Discovery in LLM Agents

Large language model (LLM) agents increasingly rely on agent harnesses that manage context, tools, and multi-turn execution, making tools a central interface for acting in realistic digital environments. As harness-connected tool ecosystems expand to hundreds or thousands of APIs, services, and task-specific skills, exhaustive tool schema injection becomes costly and imposes a closed-world assumption that limits agents to a predefined static inventory. Retrieval-augmented tool selection offers a natural alternative, but existing one-shot retrieval methods often fail to align isolated tool descriptions with the agent's true task intention, especially in long-horizon tasks where required capabilities emerge through decomposition, observations, and newly induced subgoals. We propose SING, an intention-aware active tool discovery framework that builds an intention-tool graph linking user intentions, tool capabilities, and tool collaboration patterns, and dynamically retrieves tools according to evolving task states. Using a unified corpus of 7,471 tools, we evaluate SING on three real-world tool-use benchmarks. SING improves Global Recall@5 by up to 59.8% and downstream success rate by up to 28.9% over baselines, while reducing full-corpus tool-schema exposure by 99.8%, demonstrating that intention-aware graph structure enables more accurate and context-efficient tool discovery in large-scale agentic ecosystems.

13.
arXiv (quant-ph) 2026-06-12

Statistical Mechanics and Symmetries of Non-Abelian Anyon Proliferation: From Deformation to Decoherence

arXiv:2606.12527v1 Announce Type: new Abstract: Topological quantum computation relies on braiding non-Abelian anyons, but requires the underlying topological order to survive imperfect state preparation and environmental noise. We show that the instability of topological order to wavefunction deformations and to decoherence, with the latter probed by syndrome distributions, are generically captured by stat-mech models whose symmetries naturally expose the corrupting anyonic excitations. As an example, we combine this framework with Monte-Carlo simulations to resolve the stability of $D_4$ topological order under deformations and quantum channels that proliferate multiple non-Abelian anyon species that individually are unable to condense. We show that beyond a finite threshold, proliferation of two non-Abelian anyon species parasitically condenses a shared Abelian-anyon fusion outcome, destroying the topological order. Our symmetry-based approach sharply differentiates the resulting trivial phase from that obtained by condensing all Abelian charges; in other words, the trivial phase "remembers" which anyons condensed. This framework provides a first step into identifying the relevant symmetry for optimal decoders, conditioned on syndrome measurements, of non-Abelian topological order.

14.
arXiv (CS.CL) 2026-06-12

SENTINEL: Failure-Driven Reinforcement Learning for Training Tool-Using Language Model Agents

Language model agents are increasingly effective in solving realistic tasks through multi-turn tool use. However, training reliable tool-using agents remains challenging in practice. While reinforcement learning provides an on-policy paradigm for improving agents from their own environment interactions, its effectiveness depends heavily on the training task distribution. When tasks are fixed before training, the task distribution can become increasingly mismatched with the policy's evolving capabilities, causing many rollouts to be spent on uninformative tasks. We propose SENTINEL, a failure-driven reinforcement learning framework that turns the Solver's rollout failures into targeted training tasks. SENTINEL follows a Controller–Proposer–Solver loop: the Controller analyzes failed trajectories and summarizes recurring error patterns, the Proposer generates executable tasks that stress these weaknesses, and the Solver is trained on the targeted tasks. On Tau2-Bench Retail with Qwen3-4B-Thinking-2507, SENTINEL improves Pass\^{}1 from 66.4 to 74.9 and outperforms RL on general synthetic tasks across Pass\^{}k metrics. These results demonstrate that model failures provide an effective and scalable source of targeted training signal for improving tool-using language model agents.

15.
arXiv (CS.CV) 2026-06-15

Gefen: Optimized Stochastic Optimizer

AdamW is a default optimizer for modern deep learning, but its first and second moment states add roughly two parameter-sized buffers to training memory. We propose Gefen, a memory-efficient optimizer that automatically shares second-moment estimates across parameter blocks and quantizes the first moment using a learned codebook, thereby reducing AdamW's memory footprint by ~8x while maintaining the same performance, corresponding to a reduction of 6.5 GiB per billion parameters. The method is motivated by a theoretical result showing that large mixed Hessian entries constrain the ratio of squared gradients toward one, suggesting that Hessian-aligned parameters are natural candidates for sharing second-moment statistics. Since computing Hessians is impractical at scale, Gefen infers block structure from the initial squared gradients, requiring no architecture-specific metadata or hyperparameters beyond AdamW defaults. Gefen learns an exact histogram-based dynamic-programming quantization codebook and reuses the same blocks for first-moment scaling. Across diverse experiments, Gefen achieves the lowest peak optimizer memory among the compared AdamW-like methods while maintaining AdamW-level performance. In FSDP and DDP training, the reduced memory footprint enables larger microbatches and improves throughput significantly over AdamW, providing a practical drop-in replacement with lower memory usage that can increase throughput and enable training larger models or using larger batch sizes. We provide the complete Python implementation, including fused CUDA kernels at https://github.com/ndvbd/Gefen

16.
arXiv (CS.AI) 2026-06-16

Orcheo: A Modular Full-Stack Platform for Conversational Search

arXiv:2602.14710v2 Announce Type: replace-cross Abstract: Conversational search (CS) requires a complex software engineering pipeline that integrates query reformulation, ranking, and response generation. CS researchers currently face two barriers: the lack of a unified framework for efficiently sharing contributions with the community, and the difficulty of deploying end-to-end prototypes needed for user evaluation. We introduce Orcheo, an open-source platform designed to bridge this gap. Orcheo offers three key advantages: (i) A modular architecture promotes component reuse through single-file node modules, facilitating sharing and reproducibility in CS research; (ii) Production-ready infrastructure bridges the prototype-to-system gap via dual execution modes, secure credential management, and execution telemetry, with built-in AI coding support that lowers the learning curve; (iii) Starter-kit assets include 45+ off-the-shelf components for query understanding, ranking, and response generation, enabling the rapid bootstrapping of complete CS pipelines. We describe the framework architecture and validate Orcheo's utility through case studies that highlight modularity and ease of use. Orcheo is released as open source under the MIT License at https://github.com/AI-Colleagues/orcheo.

17.
arXiv (CS.AI) 2026-06-12

Reasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday Reasoning

arXiv:2606.13607v1 Announce Type: new Abstract: When large language models (LLMs) fail to generalize or make haphazard errors in reasoning, it is often taken as evidence that LLMs are not truly reasoning, but rather performing a kind of pattern matching. The implication is that people's behavior does not exhibit the same types of failures because human reasoning uses principled and abstract world models. We evaluate human participants and 25 LLMs on their ability to engage in common-sense reasoning about a variety of everyday situations and observe similar patterns of errors in both people and models. We then identify the set of attention heads driving LLM responses and find that these heads implement a form of pattern-matching. These attention heads allow us to predict seemingly inexplicable reasoning errors in people caused by ostensibly irrelevant prompt details. Taken together, our results suggest that everyday causal reasoning in people and LLMs is more consistent with a form of pattern-matching than with abstract world models.

18.
arXiv (CS.AI) 2026-06-17

STAR: SpatioTemporal Adaptive Reward Allocation for Text-to-Image RL Post-Training

arXiv:2606.17979v1 Announce Type: new Abstract: Existing RL post-training methods for text-to-image generation usually convert the final-image reward into a single scalar advantage and apply it with the same strength to the entire generative trajectory. However, text-to-image generation naturally has temporal and spatial structure: different denoising steps are responsible for different generation stages, and the content that truly determines text alignment often appears only in part of the image. This granularity mismatch makes it difficult for policy updates to focus on the generative components that actually affect the reward. To address this issue, we propose SpatioTemporal Adaptive Reward (STAR) Allocation for RL post-training of text-to-image diffusion and flow models. STAR uses text-image attention inside the generative model and starts from the core content that the user truly cares about in the prompt. It constructs spatial allocation maps that dynamically vary across denoising steps and rollouts, and allocates the same group-relative advantage to more relevant latent regions with almost no additional computational overhead. STAR then applies stronger policy updates to these regions through a spatially resolved policy objective. We use Stable Diffusion 3.5 Medium as the base model and evaluate on three tasks: GenEval, OCR text rendering, and PickScore. Experimental results show that STAR improves compositional semantic alignment, text rendering, and preference optimization without changing the external reward source, achieving $\mathbf{0.9759}$, $\mathbf{0.9757}$, and $\mathbf{23.60}$ on GenEval, OCR, and PickScore, respectively.

19.
arXiv (CS.LG) 2026-06-12

A unified complexity bound for logconcave sampling

arXiv:2606.12694v1 Announce Type: cross Abstract: We give a simple, unified, and nearly tight bound for sampling arbitrary logconcave distributions from a warm start using the In-and-Out algorithm along with exponential lifting. The main new ingredient in the analysis is an improved bound on the Poincaré constant of a lifted distribution. As a consequence, the resulting convergence rate is nearly tight for both constrained settings (e.g., Gaussian restricted to a convex body) and well-conditioned settings (e.g., strongly logconcave and smooth densities).

20.
arXiv (CS.LG) 2026-06-17

A Generalized Sinkhorn Algorithm for Mean-Field Schrödinger Bridge

arXiv:2604.06531v3 Announce Type: replace-cross Abstract: The mean-field Schrödinger bridge (MFSB) problem concerns designing a minimum-effort controller that guides a diffusion process with nonlocal interaction to reach a given distribution from another by a fixed deadline. Unlike the standard Schrödinger bridge, the dynamical constraint for MFSB is the mean-field limit of a population of interacting agents with controls. It serves as a natural model for large-scale multi-agent systems. The MFSB is computationally challenging because the nonlocal interaction makes the problem nonconvex. We propose a generalization of the Hopf-Cole transform for MFSB and, building on it, design a Sinkhorn-type recursive algorithm to solve the associated system of integro-PDEs. Under mild assumptions on the interaction potential, we discuss convergence guarantees for the proposed algorithm. We present numerical examples with repulsive and attractive interactions to illustrate the theoretical contributions.

21.
arXiv (CS.LG) 2026-06-19

Influence-Guided Concolic Testing of Transformer Robustness

arXiv:2509.23806v2 Announce Type: replace-cross Abstract: Concolic testing for neural networks alternates concrete execution with constraint solving to search for inputs that flip model decisions. We present a concolic tester for Transformer classifiers that uses SHAP estimates to rank pending path predicates by their impact on the current prediction. To support self-attention with multiple heads in execution backed by SMT solving, we implement attention semantics in pure Python that are compatible with the solver and make the softmax boundary explicit by concretizing exponentiation arguments. We evaluate our method on CIFAR-10 across three compact Transformer classifiers, ResNet18, and VGG16 under a one-pixel budget and a 900s horizon. Across the 500 model–input pairs in this matched comparison, our method achieves 60% success, compared with 15% for a differential evolution baseline that treats the model as a black box. In the primary two-layer Transformer branch-ordering study, SHAP-based predicate prioritization raises success from 56% to 60% and reduces median attack time by 51%. These results show that influence-guided path exploration can make concolic testing a practical way to find adversarial examples in Transformer models.

22.
arXiv (CS.CV) 2026-06-11

CFCamo: A Counterfactual Detect-or-Abstain Framework for Camouflaged Object Detection

Vision-language reinforcement learning has recently shown strong target-present localization for camouflaged object detection (COD). Yet localization is only one side of the decision: when the agent faces an ordinary image with no camouflaged target, will it still claim that a camouflaged object exists? Standard COD training and evaluation data are positive-only, so agents optimized under this setting can acquire an over-detect bias, a task-specific form of object hallucination that standard COD evaluation leaves unmeasured. To quantify this target-absent behavior, we construct Counterfactual COD (CF-COD), a paired benchmark that removes the camouflaged target from each held-out COD evaluation image while preserving a plausible background. CF-COD evaluates whether a model detects the target on the original image and abstains on the target-absent counterfactual, summarized by Pair Accuracy (PA). We further introduce CFCamo, a paired counterfactual framework for COD with abstention. For training, CFCamo optimizes a Qwen3-VL-4B-Instruct agent with Counterfactual Sequence Policy Optimization (CSPO), which samples paired original-counterfactual rollouts and uses a Counterfactual Paired Reward (CPR) to couple original-image detection with counterfactual abstention. On CAMO-test, CFCamo improves S_alpha by +3.7 pp over the prior RL-based COD baseline; across CF-COD, it reaches 80.0-90.8% PA. Ablations show that removing counterfactual coupling reduces PA to 1.4-5.2% despite strong target-present COD scores, showing that target-present evaluation alone does not characterize detect-or-abstain behavior. Overall, these results indicate that CFCamo improves COD agents by coupling target-present detection with target-absent abstention, rather than merely strengthening target-present localization. Code and data are available at https://github.com/suhang2000/CFCamo.

23.
arXiv (quant-ph) 2026-06-16

Morphology-resolved scrambling in a chaotic quantum billiard

arXiv:2606.16865v1 Announce Type: new Abstract: Chaotic quantum systems can retain spatial memory through scarred eigenstates, but whether these static structures control scrambling remains unclear. This work establishes a morphology-resolved connection between scarred eigenstates and eigenstate-resolved OTOCs in a peanut-shaped quantum billiard. Scalar localisation diagnostics, including differential entropy and continuum participation ratios, detect anomalous concentration but discard spatial architecture. A scale-normalised density overlap, in contrast, directly compares probability density profiles, revealing families of orthogonal eigenstates with nearly identical spatial morphology. Comparing the complete OTOC time traces of these orthogonal eigenstates reveals that morphological recurrence has dynamical content: moderate density overlap yields no universal prediction, whereas strongly recurring morphologies exhibit nearly identical OTOC growth and saturation. Thus, scarred structures act as spatial templates for operator growth, not merely static violations of ergodicity. This morphology-resolved framework turns eigenstate shape into a quantitative predictor of scrambling and provides a scale-controlled diagnostic of weak ergodicity breaking in quantum chaos.

24.
arXiv (CS.CL) 2026-06-15

LLMs Contain Multitudes: How Deployment Context Reshapes Model-Level Preferences and Values

Large language models (LLMs) are increasingly characterised in recent evaluation work as having stable, model-level preference and value systems. However, accompanying robustness checks are limited to incidental prompt perturbations such as syntax variation and option reordering. This leaves open whether the measured properties survive when the surrounding task context changes, as it does in most real deployments. We test this directly across two established pairwise paradigms: ranking country preferences and eliciting utility judgements. In both, we make the deployment context – the high-level task the model is performing while making concrete value-dependent choices – our controlled variable, varied across framings such as writing a Reddit post or a news article. Across five LLMs and over 1.2M pairwise decisions, deployment context produces variation far larger than prompt paraphrasing and temperature controls. In country preference rankings over 15 countries, context induces widespread, statistically significant rank shifts; the aggregate Global North favouritism reported in prior work is itself context-dependent, with each model's bias shifting systematically across contexts. In utility elicitation over 50 outcomes, broad cross-category ordering is preserved, but fine-grained rankings within domains vary substantially, and cardinal exchange rates between outcomes (e.g. how many lives in one region equal one in another) shift by a factor of 2.47 at the median. Reported model-level preferences and utilities are therefore better understood as context-conditioned measurements than fixed model-level properties: safety guarantees obtained under one framing provide limited assurance in another.

25.
arXiv (CS.CV) 2026-06-17

TerraTransfer: Learning End-to-End Driving Policies Without Expert Demonstrations

End-to-end autonomous driving has achieved state-of-the-art performance on benchmarks and real-world deployments. Its standard training recipe, however, is expensive across all stages: collecting and labeling millions of driving frames is costly, and closed-loop RL on images is bottlenecked by the per-step cost of photorealistic rendering plus a forward pass through a large vision backbone. Self-play in vectorized simulators changes the economics: millions of rollout steps per second, and a state distribution naturally rich in collisions, near-misses, and recoveries that no driving log contains. Our approach exploits this asymmetry by decoupling learning to drive from learning to see. We pretrain a single policy by self-play, then align its latent space with a pretrained vision backbone, through the action KL divergence and a batch-relational low-rank structural loss. The action target comes from the self-play policy, so alignment never supervises against a logged trajectory: a paired dataset of (image, scene-state) frames suffices, with no need for the curated expert demonstrations that imitation pretraining is built on. On photorealistic 3D Gaussian splatting closed-loop scenarios, the resulting end-to-end policy matches or exceeds prior end-to-end methods.