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.CL) 2026-06-18

LLMZero: Discovering Adaptive Training Strategies for RL Post-Training via LLM Agents

RL post-training strategies are dataset-dependent and reveal a recurring empirical pattern: capacity parameters accumulate monotonically across stages, while regularization parameters predominantly oscillate in response to shifting training dynamics. This distinction matters because fixed schedules commit all parameters to fixed trajectories and therefore cannot express the non-stationary exploration-exploitation tradeoffs that regularization must track; the principle provides actionable design rules for multi-stage training. We discover this through LLMZero, a system where LLM agents search over training trajectories via tree search, diagnosing pathologies at each checkpoint and proposing coordinated multi-parameter transitions. Across 4 diverse GRPO tasks, LLMZero discovers strategies that improve over the base model by 9% to 140% relative and over grid search by 6% to 15% relative, consistently outperforming random search and the skill-based agent. The structural principle transfers across tasks, providing an explanation for why discovered strategies take qualitatively different forms yet share similar parameter dynamics.

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

Phase-Aware Guidance Injection for Recurrent MAPPO in Assembly-Line Disruption Recovery

arXiv:2606.16330v1 Announce Type: new Abstract: Disruption recovery in industrial assembly lines requires timely decisions under machine faults, worker absence, and emergency orders. Existing methods either rely on rigid handcrafted recovery logic or learn adaptive policies that do not readily exploit heterogeneous external recovery knowledge at decision time to reduce abnormal recovery time (ART) and preserve on-time delivery (OTD). To address this gap, we propose a phase-aware guidance injection framework that augments a trained recurrent MAPPO (RMAPPO) scheduling policy through logit-level action bias during evaluation. The framework provides a unified decision-time interface for rule-based, replay-based, and online LLM-based guidance, while activating intervention only during abnormal and recovery phases. Experiments on a custom AssemblyLineEnv show that high-quality rule guidance yields the strongest gains, replay-based guidance degrades smoothly under imperfect availability, and online LLM guidance still provides useful intermediate improvements. These results show that decision-time guidance injection can exploit heterogeneous recovery hints without redesigning the actor.

03.
arXiv (quant-ph) 2026-06-16

Symmetry Breaking through Superselection by Boundary Conditions

arXiv:2606.15272v1 Announce Type: cross Abstract: Spontaneous symmetry breaking (SSB) is central to modern physics but is conventionally defined only for infinite systems, raising challenges for its interpretation in finite, real-world setups. This paper argues that the key to resolving this issue lies in the underappreciated role of boundary conditions in quantum systems. Inspired by both the relational approach to symmetries and the physical mechanism behind symmetry breaking, we formulate a relational interpretation of SSB: a finite system exhibits SSB relative to a reference environment which can induce perturbations across the boundary. This eliminates the need for the thermodynamic limit, offering a more physical picture of SSB that emphasizes the observable consequences of the interactions that real-life systems inevitably have with their environment. We show how, in this relational interpretation, SSB for both lattice systems and (gauge) field theories should be understood as subtle, rather than spontaneous, symmetry breaking, still in contrast to explicit symmetry breaking. We also explain how algebraic definitions of SSB for infinite systems relate to the intuitive picture of SSB in finite systems and illustrate how asymptotic boundary conditions push the environment "to infinity". In this way, our relational interpretation of SSB provides a unified conceptual framework applicable to symmetry-breaking in systems of any size.

04.
arXiv (CS.LG) 2026-06-12

Single vs. Multiple Branches in DeepONet and S-DeepONet: Network Architecture Follows Coupling in Multiphysics Systems

arXiv:2507.03660v2 Announce Type: replace Abstract: `Real-time prediction of complex physical systems requires surrogate models that learn from data while representing strong multiphysics coupling. Deep Operator Networks have shown success in single-physics problems, yet their effectiveness in capturing nonlinear interactions in coupled systems (such as thermo-mechanical or electro-thermal coupling) remains underexplored. Here we pose a practical question: should the architecture of a neural operator reflect the strength of physical coupling it aims to model? We compare single-branch and multi-branch designs, in both feedforward and sequential recurrent forms, across three representative systems: a reaction–diffusion problem with heterogeneous sources, a nonlinear thermo-electrical problem with temperature-dependent conductivity and Joule heating, and a viscoplastic thermo-mechanical model of steel solidification. Single-branch networks consistently outperform multi-branch variants in tightly coupled regimes by encouraging shared latent representations, whereas multi-branch designs remain favorable for decoupled or single-physics tasks. Once trained, these surrogates deliver full-field predictions up to $1.8 \times 10^4$ times faster than physics-based solvers.

05.
arXiv (quant-ph) 2026-06-25

Evolving Quantum Error-Correcting Encodings for Molecular Simulation

arXiv:2606.25870v1 Announce Type: new Abstract: Useful quantum algorithms require many coupled discrete design choices. We study LLM-driven evolutionary program synthesis – a language model edits a program, an external verifier scores the result, and high-scoring programs are retained and re-mutated – as a tool for quantum-computing research. As a case study, we apply this loop to the Generalized Superfast Encoding (GSE), a fermion-to-qubit encoding whose prior molecular constructions reach code distance $3$. The search discovered interpretable constructor programs whose codes have exact distance $5$ on the molecular instances tested, and distance $6$ on one $20$-mode instance, under strict stabilizer-coset semantics. To our knowledge these are the first GSE/superfast encodings beyond distance $3$ for dense molecular Hamiltonians. A second search, guided by verifier analysis of the first artifact, found a circulant constructor that reaches a five-qubits-per-mode floor on the tested $12$-, $14$-, $16$-, and $20$-mode instances, with certified dense-rule fallback at the failing $18$-mode case. As secondary resource descriptors, in a code-capacity memory comparison at $p=10^{-3}$ the resulting encodings use $4.2$–$5.0\times$ fewer data qubits than a scoped per-mode Jordan–Wigner $+$ $[[25,1,5]]$ surface route and have $3.4$–$8.2\times$ lower logical-failure rates under finite-weight decoding tables with explicit truncation brackets; we claim no circuit-level fault-tolerance or Trotter-cost advantage. The search trajectory illustrates a general operating lesson: rewarding distance alone selects trivial dense graphs, whereas holding verified distance fixed and rewarding compression selects structured rules.

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

IntElicit: Eliciting and Assessing Contextualized Creativity via Dialogue Policy Optimization

arXiv:2606.12086v1 Announce Type: new Abstract: Contextualized assessment offers high ecological validity for evaluating creativity but introduces a critical challenge: observed performance may be confounded with cognitive proficiency (domain knowledge) and agency (willingness to engage). Meanwhile, in the age of generative AI, creative problem solving increasingly occurs in tool-mediated and human–AI interactive environments, making fully static assessment less aligned with contemporary creative practice. To address these issues, this paper proposes IntElicit, a framework for eliciting and assessing contextualized creativity via dialogue policy optimization. IntElicit functions as a constrained adaptive AI Interviewer: it provides non-directive knowledge and agency scaffolds in multi-turn interaction to reduce non-creative confounders, while preserving participants' responsibility for generating the creative content being evaluated. Specifically, to tackle sparse rewards and potential reward hacking (e.g., answer dictation) in open-ended educational dialogue, IntElicit introduces a decomposed process reward mechanism. This mechanism aligns the policy with pedagogical elicitation, rewarding prompts that draw out participant reasoning rather than producing optimal answers on their behalf. Extensive experiments, including participant simulation and a human subject study (N=64), show that IntElicit improves elicited creative outcomes over expert-designed baselines. Together, the results suggest that interactive elicitation can reveal creative potential that static FPSP-style assessment may miss, providing a formative and diagnostic lens for contextualized creativity assessment in AI-mediated learning contexts.

07.
arXiv (CS.AI) 2026-06-25

WinDOM: Self-Family Distillation for Small-Model GUI Grounding

arXiv:2606.25964v1 Announce Type: new Abstract: Small ($\sim$2B) GUI-grounding agents are attractive for on-device deployment, accessibility tooling, and low-cost iteration, but at this scale they face two open recipe questions: how to obtain bounding-box training data without expensive human annotation, and how to combine supervised fine-tuning with reinforcement learning. We address both, with the explicit goal of pushing small-model performance rather than scaling up. WinDOM is a $54{,}425$-record grounding corpus harvested by driving an open-source Windows 11 web reimplementation under headless Playwright, with bounding boxes read directly off the DOM and no OCR or human annotation. Self-Family Distillation (SFD) is a single rejection-sampling cold-start parameterised only by the teacher choice: either an EMA of the student (no external model) or a frozen larger same-family teacher. We then treat the saturation depth of the SFD cold-start as an explicit GRPO hyperparameter. On a Qwen3.5-2B student, the under-saturated cold-start is a better GRPO initialiser than the converged one: SFD-4B with Early-init RL gains $+5.4$ OOD-mean ($+3.5$ ScreenSpot-Pro, $+7.0$ OSWorld-G, $+5.8$ ScreenSpot-V2) over the base. The same-size EMA mode lands within roughly one OOD-mean point of the cross-size $4$B variant ($65.2$ vs $66.3$) without an external teacher.

08.
arXiv (CS.CL) 2026-06-18

JetFlow: Breaking the Scaling Ceiling of Speculative Decoding with Parallel Tree Drafting

Speculative decoding (SD) accelerates autoregressive Large Language Models (LLMs) by drafting multiple tokens and verifying them in parallel, but it faces a scaling limitation: increasing the draft budget improves speed only when acceptance remains high and drafting overhead stays low. This ceiling has been difficult to break because prior head-based SD methods face a causality-efficiency dilemma. Autoregressive drafters produce path-conditioned candidates that are effective for tree speculative decoding with higher acceptance length, but their drafting cost grows with tree depth. Bidirectional block-diffusion drafters generate all positions in one pass, but their branch-agnostic marginals can form individually plausible yet mutually inconsistent trees, wasting budget and reducing acceptance. We propose JetFlow, a head-based SD framework that combines one-forward drafting efficiency with branch-wise causal conditioning. JetFlow trains a causal parallel draft head over fused hidden states from the frozen target model, producing candidate trees whose scores align with the target model's autoregressive factorization. This enables JetFlow to convert larger draft budgets into longer accepted prefixes and higher end-to-end speedup. Across math, coding, and chat benchmarks on dense and MoE Qwen3 models, JetFlow consistently outperforms bidirectional-head and tree-based SD baselines. On H100 GPUs, JetFlow achieves up to 9.64x speedup on MATH-500 and 4.58x on open-ended conversational workloads, with further latency gains demonstrated through vLLM integration under realistic serving loads. Our code and models are available at https://github.com/hao-ai-lab/JetFlow.

09.
arXiv (CS.LG) 2026-06-16

Probing Dec-POMDP Reasoning in Cooperative MARL

arXiv:2602.20804v2 Announce Type: replace Abstract: Cooperative multi-agent reinforcement learning (MARL) is typically framed as a decentralised partially observable Markov decision process (Dec-POMDP), a setting whose hardness stems from two key challenges: partial observability and decentralised coordination. Genuinely solving such tasks requires Dec-POMDP reasoning, where agents use history to infer hidden states and coordinate based on local information. Yet it remains unclear whether popular benchmarks actually demand this reasoning or permit success via simpler strategies. We introduce a diagnostic suite combining statistically grounded performance comparisons and information-theoretic probes to audit the behavioural complexity of baseline policies (IPPO and MAPPO) across 37 scenarios spanning MPE, SMAX, Overcooked, Hanabi, and MaBrax. Our diagnostics reveal that success on these benchmarks rarely requires genuine Dec-POMDP reasoning. Reactive policies match the performance of memory-based agents in over half the scenarios, and emergent coordination frequently relies on brittle, synchronous action coupling rather than robust temporal influence. These findings suggest that some widely used benchmarks may not adequately test core Dec-POMDP assumptions under current training paradigms, potentially leading to over-optimistic assessments of progress. We release our diagnostic tooling to support more rigorous environment design and evaluation in cooperative MARL.

10.
arXiv (math.PR) 2026-06-11

The $K$-th nearest neighbor random walk on a Poisson point process gets trapped

arXiv:2606.11271v1 Announce Type: new Abstract: The $K$-th nearest neighbor random walk $(X_n)_{n \geq 0}$ on a homogeneous Poisson point process $\chi$ on $\R^d$ ($d\geq 1$), starts at the origin and at each step picks its next Poisson point among its closest neighbors according to i.i.d. labels having the same distribution as $K$. Our main result (Theorem 1) states that the number of Poisson points visited by $(X_n)_{n \geq 0}$ admits an exponential decay whenever the random variable $K$ has a bounded support (BS). In particular, the $K$-th nearest neighbor random walk visits finitely many Poisson points if and only if $K$ satisfies Assumption (BS). To prove it, we introduce the key notion of pioneer point which allows us to deal with the region of $\R^d$ already explored by $(X_n)_{n \geq 0}$. Still under Assumption (BS), we also prove an exponential decay for the Euclidean length of the trajectory performed by $(X_n)_{n \geq 0}$ (Theorem 2). Finally, and quite surprisingly, we exhibit an example of label distribution with bounded support for which the $K$-th nearest neighbor random walk discovers new Poisson points after a number of steps whose tail distribution is at least polynomial (Theorem 3).

11.
medRxiv (Medicine) 2026-06-12

High coverage, persistent gaps: quality of Antenatal Care and its determinants in Zambia based on the 2024 Demographic and Health Survey.

Abstract Background Evaluating antenatal care (ANC) quality is critical to reducing maternal and neonatal mortality. In Zambia, despite high basic ANC attendance, comprehensive national evidence on the clinical content and quality of services remains limited. This study assessed the coverage of WHO-recommended ANC interventions and identified factors associated with care quality using the latest national data. Methods A cross-sectional analysis was conducted using data from the 2024 Zambia Demographic and Health Survey. The final analytic sample comprised 4,829 women aged 15-49 with a live birth in the preceding 5 years. A composite index of 15 selected, equally weighted WHO-recommended components evaluated clinical assessment, counseling/screening, preventive interventions, and utilization. Survey-weighted Poisson regression estimated adjusted incidence rate ratios (aIRRs) for the count of ANC components received. Results The mean ANC quality score was 12.5 out of 15 (95% CI: 12.4-12.6), and 78.5% (95% CI: 77.0-80.0) of women achieved adequate ANC ([≥] 12/15 components). While individual clinical and counseling coverage generally exceeded 90%, only 47.2% (95% CI: 45.3-49.0) of women initiated care during the first trimester, and just 4.8% (95% CI: 4.1-5.6) achieved [≥] 8 ANC contacts. Maternal education was the strongest and most stable predictor of quality across all models. Compared to no education, higher education was associated with an 8.0% higher expected quality score (aIRR = 1.080, 95% CI: 1.051-1.110). Lower ANC quality was significantly associated with unwanted pregnancies (aIRR = 0.970, 95% CI: 0.956-0.993) and with residence in Western (aIRR = 0.923, 95% CI: 0.897-0.951) and North Western (aIRR = 0.966, 95% CI: 0.937-0.996) provinces. Absence of distance barriers and residence in Eastern, Luapula, and Copperbelt provinces were associated with higher quality scores. Conclusion While average ANC component coverage in Zambia is high, critical gaps persist in early initiation and total contact frequency. Care adequacy is strongly influenced by maternal education, relationship status, pregnancy intention, and regional inequities. These findings underscore the need for interventions targeted at uneducated women, preventing unintended pregnancies, and underserved regions such as Western and North Western Provinces. Keywords: Antenatal care quality, ANC content, Zambia, maternal education.

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

Exploiting Search in Symbolic Numeric Planning with Patterns

arXiv:2606.16329v1 Announce Type: new Abstract: In this paper, we present a procedure for numeric planning based on Symbolic Pattern Planning (SPP). Given a numeric planning problem $\Pi$, a pattern $\prec$ is a sequence of actions used to define a formula encoding the subsequences of $\prec$ executable from a starting state $S$. Cardellini, Giunchiglia, and Maratea (2024a) follow the Planning as Satisfiability approach by defining, at each step $n \ge 0$, a formula $\Pi^\prec_n$ in which $(i)$ the pattern $\prec$ is computed only for $n=0$ in the initial state $I$ of $\Pi$, and then exploited at each step $n$, $(ii)$ the starting state $S$ is set to $I$, and $(iii)$ the set $G$ of goals is required to hold in the last state that can be reached by one of the subsequences of $\prec$ concatenated $n$ times. The procedure begins with $n=0$, terminates as soon as $\Pi^\prec_n$ is satisfiable, and otherwise proceeds by incrementing $n$. In this paper, possibly at each step, $(i)$ we symbolically search for an intermediate state $P$ reachable from $I$, closer to a goal state, $(ii)$ dynamically recompute the pattern $\prec_h$ – to be used in the next step – in $P$, $(iii)$ refine the pattern $\prec_g$ used to reach $P$, and $(iv)$ start the new search from the state $S$ which can be either the initial state $I$ or the last computed intermediate state $P$, exploiting the computed patterns $\prec_g$ and $\prec_h$ to define the pattern $\prec$ to be used in the search. In particular, at each step, we define a formula $\Pi^{\prec}_{S,P}$ encoding the existence of a state $P'$ closer than $P$ to a goal state, with $P'$ reachable from the starting state $S$ when using the pattern $\prec$. We present different techniques for producing such formulas, each corresponding to a different strategy for exploring the search space. We prove their correctness and completeness, the latter under certain conditions.

13.
arXiv (math.PR) 2026-06-25

Breuer-Major Theorems for Hilbert Space-Valued Random Variables

arXiv:2405.11452v2 Announce Type: replace Abstract: Let $\{X_k\}_{k\in\mathbb{Z}}$ be a stationary Gaussian process with values in a separable Hilbert space $\mathcal{H}_1$, and let $G:\mathcal{H}_1\to\mathcal{H}_2$ be a measurable map into another separable Hilbert space $\mathcal{H}_2$. We derive a central limit theorem for the centered normalized partial sums of the Hilbert space-valued subordinated process $\{G[X_k]\}_{k\in\mathbb{Z}}$. Our result holds under either of two sets of sufficient conditions, formulated in terms of the transformation $G$ and the temporal and cross-sectional dependence structure of $\{X_k\}_{k\in\mathbb Z}$. These conditions coincide in finite dimensions but lead to genuinely different phenomena in the infinite-dimensional setting. The proof relies on the recently developed Fourth Moment Theorem on Hilbert spaces, leveraging tools from the infinite-dimensional Malliavin-Stein framework. We also provide continuous-time and quantitative versions of the central limit theorem. In a series of examples, we recover and strengthen limit theorems for a wide array of statistics relevant in functional data analysis, and present, as an application of our result, a novel limit theorem in the framework of neural operators.

14.
arXiv (CS.AI) 2026-06-11

WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces

arXiv:2606.09426v2 Announce Type: replace Abstract: Computer-use agents (CUAs) increasingly operate in runtimes that combine visual desktop control, command-line execution, code editing, browsers, and external tools. Existing benchmarks, however, often evaluate these interfaces as separable capabilities, leaving long-horizon cross-interface orchestration under-tested. Thus, we introduce WeaveBench, a long-horizon hybrid-interface benchmark with 114 tasks across 8 real-world work domains, grounded in real user requests and publicly verifiable artifacts. Each task requires agents to combine GUI observations/actions with CLI/code operations within a single trajectory. We evaluate these tasks on a real Ubuntu desktop inside deployed CLI-agent runtimes, augmented with a minimal desktop-control plugin. We also propose a companion trajectory-aware judge that inspects deliverables, files, screenshots, logs, and action traces, while detecting shortcut behaviors such as fabricated visual evidence or hard-coded metrics. Across frontier model-runtime pairings, the best PassRate reaches only 41.2%, showing the benchmark remains far from saturated. The trajectory-aware judge further reveals that outcome-only grading substantially overestimates agent performance. Overall, WeaveBench exposes a critical gap in CUA evaluation and provides an effective testbed to measure whether agents can orchestrate GUI, CLI, and code operations across long-horizon real-world tasks.

15.
arXiv (quant-ph) 2026-06-16

Intrinsic preservation of plasticity in continual quantum learning

arXiv:2511.17228v2 Announce Type: replace Abstract: Artificial intelligence in dynamic, real-world environments requires the capacity for continual learning. However, standard deep learning suffers from a fundamental issue: loss of plasticity, in which networks gradually lose their ability to learn from new data. Here we show that quantum learning models naturally overcome this limitation, preserving plasticity over long timescales. We demonstrate this advantage systematically across a broad spectrum of tasks from multiple learning paradigms, including supervised learning and reinforcement learning, and diverse data modalities, from classical high-dimensional images to quantum-native datasets. Although classical models exhibit performance degradation correlated with unbounded weight and gradient growth, quantum neural networks maintain consistent learning capabilities regardless of the data or task. We identify the origin of the advantage as the intrinsic physical constraints of quantum models. Unlike classical networks where unbounded weight growth leads to landscape ruggedness or saturation, the unitary constraints confine the optimization to a compact manifold. Our results suggest that the utility of quantum computing in machine learning extends beyond potential speedups, offering a robust pathway for building adaptive artificial intelligence and lifelong learners.

16.
arXiv (math.PR) 2026-06-16

On the empirical spectral distribution of matrix perpetuities

arXiv:2605.31054v2 Announce Type: replace Abstract: We study matrix perpetuities, that is, solutions to affine fixed-point equations of the form \[ \mathbf{X} \stackrel{d}{=} \mathbf{A}\,\mathbf{X} \,\mathbf{A}^\top+\mathbf{B},\qquad (\mathbf{A},\mathbf{B})\mbox{ and }\mathbf{X} \mbox{ are independent}, \] with particular emphasis on the empirical spectral distribution of the solution. We first establish existence and uniqueness results by relating the problem to classical vector perpetuities, and then develop tools that preserve the matrix structure under orthogonal invariance. For positive semidefinite, orthogonally invariant models, we obtain power-law tail asymptotics for the expected empirical spectral distribution and show that the tail is governed by the largest eigenvalue. We also prove that, in the subcritical regime, the expected empirical spectral distribution of matrix perpetuities converges weakly, as the dimension tends to infinity, to the distribution of the corresponding free perpetuity. Our results are illustrated by matrix Beta prime perpetuities, for which explicit limiting spectral distributions are available.

17.
arXiv (CS.LG) 2026-06-19

On the Variance of Temporal Difference Learning and its Reduction Using Control Variates

arXiv:2606.20357v1 Announce Type: new Abstract: We analyze the variance of temporal difference (TD) learning using the phased setting with tabular representation, and show that one of the mechanisms behind its ability to reduce variance is by effectively aggregating over a larger number of independent trajectories. Based on this insight, we demonstrate that (1) the variance of TD is asymptotically bounded from above by Monte Carlo (MC) estimators, and (2) shorter horizon updates incurs less variance for a fixed number of samples. Beyond TD, we show that Direct Advantage Estimation (DAE), a method for estimating the advantage function, can be seen as a type of regression-adjusted control variate, which achieves a tighter bound on the variance compared to TD in the large-sample limit. Finally, we numerically illustrate the behaviors of these estimators with carefully designed environments.

18.
arXiv (CS.AI) 2026-06-16

Sample from What You See: Visuomotor Policy Learning via Diffusion Bridge with Observation-Embedded Stochastic Differential Equation

arXiv:2512.07212v3 Announce Type: replace Abstract: Imitation learning with diffusion models has advanced robotic control by capturing the multi-modal action distributions. However, existing methods typically treat observations only as high-level conditions to the denoising network, rather than integrating them into the stochastic dynamics of the diffusion process itself. As a result, the sampling is forced to begin from random noise, weakening the coupling between perception and control and often yielding suboptimal performance. We propose BridgePolicy, a generative visuomotor policy that directly integrates observations into the stochastic dynamics via a diffusion-bridge formulation. By constructing an observation-informed trajectory, BridgePolicy enables sampling to start from a rich and informative prior rather than random noise, substantially improving precision and reliability in control. A key difficulty is that diffusion bridge normally connects distributions of matched dimensionality, while robotic observations are heterogeneous and not naturally aligned with actions. To overcome this, we introduce a semantic aligner to unify the visual and state inputs and align the observations with action representations, making diffusion bridge applicable to heterogeneous robot data. Extensive experiments across 52 simulation tasks on three benchmarks and 5 real-world tasks demonstrate that BridgePolicy consistently outperforms state-of-the-art generative policies. Our code is available at https://jianghcsr.github.io/BridgePolicy_page/.

19.
arXiv (CS.LG) 2026-06-16

Evolutionary Bilevel Reward Shaping for Generalization in Reinforcement Learning

arXiv:2606.16236v1 Announce Type: new Abstract: Reinforcement learning (RL) often suffers from performance degradation when deployed in environments that differ from those encountered during training. Existing techniques such as domain randomization (DR) mitigate this, but require access to diverse training environments and full trajectory observability, assumptions that fail in privacy-preserving or restricted scenarios where only scalar performance metrics are available. We propose Generalization via Evolutionary Reward Shaping (GERS), a bilevel optimization approach to improve generalization on unseen test environments using only scalar feedback from validation environments. At the lower level, an RL agent guided via a reward function shaped by the upper level learns a policy on a limited set of training environments with accessible trajectory data; at the upper level, CMA-ES optimizes the reward shaping parameters to maximize the cumulative unshaped reward on separate validation environments for which trajectory access is unavailable. Results on continuous control tasks indicate that GERS outperforms the standard RL baseline on unseen test environments. GERS performance is comparable to DR, despite DR treating the combined set of training and validation environments of GERS as a single training set that requires trajectory access, whereas GERS cannot access validation trajectories. These results confirm that GERS effectively enhances generalization under restricted data access constraints.

20.
arXiv (CS.CV) 2026-06-24

VisChronos: Revolutionizing Image Captioning Through Real-Life Events

This paper aims to bridge the semantic gap between visual content and natural language understanding by leveraging historical events in the real world as a source of knowledge for caption generation. We propose VisChronos, a novel framework that utilizes large language models and dense captioning models to identify and describe real-life events from a single input image. Our framework can automatically generate detailed and context-aware event descriptions, enhancing the descriptive quality and contextual relevance of generated captions to address the limitations of traditional methods in capturing contextual narratives. Furthermore, we introduce a new dataset, EventCap (https://zenodo.org/records/14004909), specifically constructed using the proposed framework, designed to enhance the model's ability to identify and understand complex events. The user study demonstrates the efficacy of our solution in generating accurate, coherent, and event-focused descriptions, paving the way for future research in event-centric image understanding.

21.
arXiv (CS.CL) 2026-06-11

"Do Not Mention This to the User": Detecting and Understanding Malicious Agent Skills in the Wild

LLM-based coding agents increasingly rely on third-party extensions called skills, which bundle natural language instructions and helper scripts that execute with full user privileges. Community registries have emerged to distribute these skills, but the security implications remain unstudied due to the absence of labeled threat data. This paper presents a systematic security analysis of 98,380 skills collected from two major registries. Through a combination of static pattern matching and dynamic behavioral verification, we identify 157 skills exhibiting confirmed malicious behavior, encompassing 632 distinct vulnerabilities across 13 attack techniques. Our analysis reveals that these threats are deliberate rather than accidental: each malicious skill contains an average of 4.03 vulnerabilities spanning multiple attack phases. We identify two dominant attack strategies with statistically significant negative correlation – credential theft via remote code execution, and agent manipulation through adversarial instructions embedded in documentation. Over half of all confirmed cases originate from a single threat actor employing templated brand impersonation at scale. We further observe that attack sophistication correlates with concealment investment, with advanced skills universally employing undocumented capabilities while also exploiting platform-native trust mechanisms. Following responsible disclosure, registry maintainers removed all 157 (100%) of the reported skills. Our dataset and detection pipeline are publicly available to facilitate future research on securing LLM agent ecosystems.

22.
arXiv (CS.CV) 2026-06-16

Metis: A Generalizable and Efficient World-Action Model for Autonomous Driving and Urban Navigation

World action models~(WAMs) have shown great promise for autonomous driving and urban navigation. Built upon Vision-Language-Action models or video generation models, existing approaches suffer key limitations: (1) High inference latency due to future observation prediction at test time, and (2) tightly coupled video and action modeling leading to representational mismatch and degraded generalization. To address both issues, we propose Metis, an end-to-end WAM framework that decouples video generation and action prediction. Specifically, Metis employs a Mixture-of-Transformers architecture with dedicated experts for video generation and action prediction, preserving the intrinsic distributional properties of each task. To enhance efficiency, we introduce an asymmetric attention mask that enables joint training of both experts while allowing the action model to bypass explicit video generation during inference. This design ensures training-inference consistency and significantly reduces computational costs without compromising planning performance. Extensive experiments demonstrate state-of-the-art performance on the NAVSIM navhard and navtest benchmarks and the CityWalker navigation benchmark, validating both the generalizability and efficiency across diverse tasks. Real-robot deployments further confirm the practical feasibility of our approach.

23.
arXiv (CS.AI) 2026-06-16

AdaSTORM: Scaling LLM Reasoning on Dynamic Graphs via Adaptive Spatio-Temporal Multi-Agent Collaboration

arXiv:2606.16328v1 Announce Type: new Abstract: Large Language Models (LLMs) demonstrate remarkable potential in dynamic graph reasoning, but suffer from a scaling bottleneck: current models can only handle graphs with tens of nodes, constrained by exponential reasoning overhead and finite context windows. While multi-agent systems (MAS) offer collective reasoning and topology-aware orchestration, capabilities naturally suited for graph-structured tasks, their application to dynamic graphs remains unexplored. This paper presents Scaling LLM Reasoning on Dynamic Graphs via Adaptive Spatio-Temporal Multi-Agent Collaboration (AdaSTORM), a framework that reformulates large-scale dynamic graph reasoning into two stages: (i) Adaptive Partitioning, partitioning large-scale dynamic graphs into subregions that match the model's reasoning capacity while minimizing inference cost; and (ii) Collaborative Reasoning, aligning graph partition topologies with a spatio-temporal decoupled multi-agent architecture. AdaSTORM is the first multi-agent framework tailored for dynamic graph reasoning. Extensive experiments show that AdaSTORM successfully breaks through the scaling bottleneck, scaling reasoning to thousand-node graphs with over 90% accuracy across several large-scale dynamic graph settings without external tools, significantly outperforms seven competitive baselines. Furthermore, it achieves state-of-the-art accuracy on existing benchmarks and generalizes robustly to real-world datasets. The source code is available at: https://github.com/irisorchid107/AdaSTORM/.

24.
arXiv (CS.AI) 2026-06-19

Bring My Cup! Personalizing Vision-Language-Action Models with Visual Attentive Prompting

arXiv:2512.20014v3 Announce Type: replace-cross Abstract: While Vision-Language-Action (VLA) models generalize well to generic instructions, they struggle with personalized commands such as "bring my cup," where the robot must act on one specific instance among visually similar objects. We study this setting of manipulating personal objects, in which a VLA must identify and control a user-specific object unseen during training using only a few reference images. To address this challenge, we propose Visual Attentive Prompting (VAP), a simple-yet-effective training-free perceptual adapter that equips frozen VLAs with top-down selective attention. VAP treats the reference images as a non-parametric visual memory, grounds the personal object in the scene through open-vocabulary detection and embedding-based matching, and then injects this grounding as a visual prompt by highlighting the object and rewriting the instruction. We construct two simulation benchmarks, Personalized-SIMPLER and Personalized-VLABench, and a real-world tabletop benchmark to evaluate personalized manipulation across multiple robots and tasks. Experiments show that VAP consistently outperforms generic policies and token-learning baselines in both success rate and correct-object manipulation, helping to bridge the gap between semantic understanding and instance-level control.

25.
arXiv (CS.AI) 2026-06-11

RAIL: Rethinking Auditory Intelligence in Large Audio-Language Models with a CHC-Grounded Benchmark

arXiv:2606.11260v1 Announce Type: cross Abstract: Humans process rich auditory environments through tightly integrated cognitive capabilities such as audio perception, audio reasoning, and memory. Despite recent progress in large audio-language models (LALMs) across speech understanding and multimodal audio reasoning, current evaluation paradigms remain largely task- or modality-centric, focusing on end performance while overlooking underlying auditory cognitive behaviours. This reveals a fundamental gap between how auditory cognition is understood in humans and how it is evaluated in LALMs, particularly in the lack of frameworks that operationalise cognitive principles beyond task-level metrics to systematically capture model behaviour. In this work, we introduce RAIL, a human-centric evaluation paradigm grounded in the Cattell-Horn-Carroll (CHC) cognitive framework. RAIL formalises auditory cognition into five core capabilities and develop them into structured evaluation tasks that probe how models process, retain, and integrate auditory information. We further construct a cognitively grounded benchmark with principled data curation and human-aligned evaluation protocols. Evaluating 26 state-of-the-art LALMs, we find that current models exhibit highly uneven performance across cognitive abilities. RAIL establishes a new evaluation paradigm that moves beyond task-centric benchmarking toward cognitively grounded assessment of auditory intelligence.