arXiv:2511.06278v2 Announce Type: replace Abstract: Thermoacoustic tomography (TAT) is an imaging technique based on the thermoacoustic effect, combining electromagnetic contrast and high resolution of ultrasound imaging. In TAT, a short micro- or radio wave pulse is directed to the imaged target. Energy of this pulse is absorbed depending on the dielectric parameters of the target, resulting in a spatially varying pressure distribution via the thermoacoustic effect. This pressure, known as the initial pressure distribution, propagates as acoustic waves that are measured on the boundary of the target using ultrasound sensors. In the inverse problem of TAT, the initial pressure is estimated from the measured ultrasound waves. TAT can further be extended to quantitative TAT (QTAT), where the aim is to estimate the dielectric parameters of the target from the measured ultrasound waves, utilizing a model for electromagnetic wave propagation. In this work, we study the inverse problem of QTAT, and propose an approach for simultaneous estimation of electrical conductivity and permittivity from the ultrasound waves. This problem is approached in the framework of Bayesian inverse problems, enabling incorporation of prior and noise models. The forward model describing electromagnetic and acoustic wave propagation is based on the Maxwell's equations and the acoustic wave-equation, respectively. The approach is evaluated with numerical simulations. The results show that the dielectric parameters can be estimated using the proposed approach with good precision. However, the ultrasound sensor geometry and the number of electromagnetic pulses have a significant effect on the accuracy of the estimated parameters.
Science Journals
arXiv:2512.11784v2 Announce Type: replace Abstract: Softmax attention is a central component of transformer architectures, yet its nonlinear structure poses significant challenges for theoretical analysis. We develop a unified, measure-based framework for studying single-layer softmax attention under both finite and infinite prompts. For i.i.d. Gaussian inputs, we lean on the fact that the softmax operator converges in the infinite-prompt limit to a linear operator acting on the underlying input-token measure. Building on this insight, we establish non-asymptotic concentration bounds for the output and gradient of softmax attention, quantifying how rapidly the finite-prompt model approaches its infinite-prompt counterpart, and prove that this concentration remains stable along the entire training trajectory in general in-context learning settings with sub-Gaussian tokens. In the case of in-context linear regression, we use the tractable infinite-prompt dynamics to analyze training at finite prompt length. Our results allow optimization analyses developed for linear attention to transfer directly to softmax attention when prompts are sufficiently long, showing that large-prompt softmax attention inherits the analytical structure of its linear counterpart. This, in turn, provides a principled and broadly applicable toolkit for studying the training dynamics and statistical behavior of softmax attention layers in large prompt regimes.
arXiv:2606.17216v1 Announce Type: new Abstract: We study strategic space- and time-constrained cooperation between two self-interested agents through the Intermittent Strategic Cooperation-Based Two-Agent Path Planning (IC2PP) problem, a shortest-path game on graphs in which agents navigate toward individual targets while optionally cooperating at specific nodes to reduce their own travel times. Although such cooperation can strictly benefit both agents, it is strategically fragile: agents may deviate at any point along their paths. Modeled as a 2-player game, we characterize the structure of Pure Nash Equilibrium (PNE) joint strategies in IC2PP, and show that stable cooperation must follow a highly constrained form. We further prove that at least one PNE exists in every instance of IC2PP, and present a polynomial-time algorithm for enumerating all relevant PNEs. When multiple equilibria arise, we study coordination mechanisms based on bargaining-theoretic selection concepts and empirically compare equilibrium outcomes in terms of individual travel times and social welfare.
arXiv:2606.17220v1 Announce Type: new Abstract: Legal case retrieval remains challenging due to the complexity of legal language and the need for precise lexical alignment between queries and relevant cases. Although dense retrieval models have achieved notable progress, empirical studies show that BM25 continues to serve as a strong baseline in this domain. It motivates us to propose a self-evolving framework for rule-driven query rewriting that enhances BM25 without any parameter training. The framework equips an LLM-based agent with an automatic evaluation environment, enabling it to iteratively create rewriting rules, plan validation experiments over rule combinations, and eliminate ineffective rules based on historical feedbacks. We evaluate our method on the Chinese legal case retrieval benchmark LeCaRD-v2. Experimental results demonstrate that the proposed framework outperforms non-evolutionary baselines, including human-designed rules and greedy rule selection, particularly when powered by a highcapacity core LLM. We also conduct detailed analyses to investigate the mechanisms underlying self-evolution. Our findings reveal that LLM's capabilities to leverage previous experimental results and its intrinsic knowledge of rule elimination play critical roles in refining the rule set via self-evolution.
arXiv:2606.17225v1 Announce Type: new Abstract: In the paraxial domain, a difference-frequency generation (DFG) process provides a versatile platform for exploring non-Hermitian dynamics in an all-classical interaction. This letter aims to show the existence of topologically non-trivial optical analogues of anti-Skyrmionic states in an 'anti' parity-time (anti-PT) symmetry-broken dynamical interaction in a DFG process. A practical DFG set-up, which is constituted by a periodically-poled lithium tantalate (PPLT) crystal and pumped by a visible (green) pump laser, is considered for ascertaining the non-Hermitian dynamics. In the anti-PT symmetry broken phase, a negatively phase-mismatched interaction gives rise to a topologically non-trivial pseudo-magnetization texture that could support anti-skyrmion-like quasi-particles in signal and idler modes. The analysis also reveals the existence of anti-meron like pseudo-magnetization texture for a phase-matched DFG.
arXiv:2606.17693v1 Announce Type: new Abstract: We show that existing tools for termination analysis are extremely well suited for LTL model checking of infinite state systems. To this end, we present a framework MoAT which uses the well-known automata-based approach and reduces the LTL model checking problem to fair termination. To prove or disprove fair termination, it then calls the termination tools KoAT and LoAT in the backend. Our experiments show that in this way, MoAT is on par with existing state-of-the-art tools for LTL model checking of infinite state systems.
arXiv:2606.17247v1 Announce Type: cross Abstract: This paper investigates the performance of tunable liquid lens (TLL)-assisted receivers in large-scale visible light communication (VLC) systems under random receiver orientation. A simple electrowetting-based TLL architecture is proposed, capable of dynamically steering the incident optical signal toward the photodiode receiver by adjusting the orientation of the liquid interface. The proposed architecture enhances the desired signal reception while mitigating interference from neighboring access points (APs). The spatial distribution of APs is modeled using a Mat\'ern hard-core point process, whereas receiver orientation is characterized by uniformly distributed azimuth angles and Gaussian-distributed polar angles. Furthermore, a tractable mathematical optical channel model is developed to capture the combined effects of AP/receiver locations, receiver orientation, and lens adjustment angles on the VLC channel gain. Based on this framework, three lens orientation strategies, namely best signal reception (BSR), closest LED selection, and vertical upward lens orientation, are proposed to improve system performance under dynamic receiver conditions. Using stochastic geometry tools, exact and approximate analytical expressions for the outage probability are derived for each scheme. Numerical results verify the accuracy of the developed analysis and demonstrate that the proposed TLL-assisted receiver architecture significantly improves the robustness of VLC systems under severe receiver orientation fluctuations and dense AP deployments. In particular, the BSR scheme reduces the outage probability by $57.1\%$ compared with conventional fixed-lens receivers at an AP height of $3.5$ m and AP density of $0.2~\text{m}^{-2}$. The presented analytical framework and numerical results provide useful design insights for the deployment of future TLL-assisted VLC networks.
arXiv:2606.17683v1 Announce Type: new Abstract: While large language models (LLMs) have greatly advanced the functional correctness of automated code translation systems, the runtime efficiency of translated programs has received comparatively little attention. With the waning of Moore's law, runtime efficiency has become increasingly important for program quality, alongside functional correctness. Our preliminary study reveals that LLM-translated programs often run slower than human-written ones, and this issue cannot be remedied through prompt engineering alone. Therefore, our work proposes SwiftTrans, a code translation framework comprising two key stages: (1) Multi-Perspective Exploration, where MpTranslator leverages parallel in-context learning (ICL) to generate diverse translation candidates; and (2) Difference-Aware Selection, where DiffSelector identifies the optimal candidate by explicitly comparing differences between translations. We further introduce Hierarchical Guidance for MpTranslator and Ordinal Guidance for DiffSelector, enabling LLMs to better adapt to these two core components. To support the evaluation of runtime efficiency in translated programs, we extend existing benchmarks, CodeNet and F2SBench, and introduce a new benchmark, SwiftBench. Experimental results across all three benchmarks show that SwiftTrans achieves consistent improvements in both correctness and runtime efficiency.
arXiv:2606.17279v1 Announce Type: new Abstract: Surgical video question answering requires multi-step reasoning across semantic, spatial, and temporal dimensions. Existing methods architecturally compress videos into discrete token representations and couple visual perception with reasoning. This approach fragments continuous spatial-temporal relationships and has been shown to restrict multi-step reasoning capabilities. We introduce a reinforcement learning (RL) framework that trains large language models (LLMs) to decouple perception from reasoning by operating over digital twin representations constructed from surgical foundation models. Additionally, we introduce hierarchical representations across frame, temporal window, and procedure levels with probabilistic uncertainty estimates. Finally, we propose a novel reward that combines format validation with accuracy assessment through clinical plausibility evaluation and uncertainty-aware calibration for training. To demonstrate the capabilities of this approach, we introduce REAL-Colon-Reason, a colonoscopic benchmark with 2000 question-answer pairs across three complexity levels. We achieve state-of-the-art performance on REAL-Colon-Reason and two existing surgical VideoQA benchmarks REAL-Colon-VQA and EndoVis18-VQA.
arXiv:2512.13009v2 Announce Type: replace Abstract: Reliable estimation of contact forces is crucial for ensuring safe and precise interaction of robots with unstructured environments. However, accurate sensorless force estimation remains challenging due to inherent modeling errors and complex residual dynamics and friction. To address this challenge, in this paper, we propose K-VARK (Kernelized Variance-Aware Residual Kalman filter), a novel approach that integrates a kernelized, probabilistic model of joint residual torques into an adaptive Kalman filter framework. Through Kernelized Movement Primitives trained on optimized excitation trajectories, K-VARK captures both the predictive mean and input-dependent heteroscedastic variance of residual torques, reflecting data variability and distance-to-training effects. These statistics inform a variance-aware virtual measurement update by augmenting the measurement noise covariance, while the process noise covariance adapts online via variational Bayesian optimization to handle dynamic disturbances. Experimental validation on a 6-DoF collaborative manipulator demonstrates that K-VARK achieves over 20% reduction in RMSE compared to state-of-the-art sensorless force estimation methods, yielding robust and accurate external force/torque estimation suitable for advanced tasks such as polishing and assembly.
arXiv:2606.17266v1 Announce Type: new Abstract: Production planning increasingly has to treat workforce capability as a decision variable: certifications lapse when skills are not maintained, new products require skills the current workforce does not hold, and reskilling competes for the same worker hours needed for production. Existing operations benchmarks usually treat labor as exogenous, while workforce-planning models with skills and learning are rarely released as reusable testbeds. We introduce SkillChain-Gym, a benchmark specification for reskilling-aware production-inventory control: a single-site environment with stylized worker skill-state dynamics, hard threshold certification, forgetting, and capacity-consuming training actions constrained by the same per-worker time budget as production. The benchmark includes seed-controlled disruption scenarios, three feasibility modes with projection diagnostics, deterministic replay, and metrics covering operations, resilience, capability growth, and training-access distribution. We evaluate production-only, reactive adaptive, water-filling adaptive, and static-insurance policies with budget variants over 60-shift horizons with paired statistical tests. The results are regime-dependent rather than a ranking. Training-capable policies dominate the production-only baseline, and maintenance training is necessary under forgetting even without disruptions. Among training-capable classes, adaptive training helps when bottlenecks are visible in the forecast, while a lean static cross-training plan, a deliberately favorable comparator whose structure encodes relevant skill contingencies, acts as strong insurance under surprise shocks and absenteeism. Capacity slack and the forgetting rate govern the boundary between these regimes. No policy class dominates across regimes, motivating forecast-driven controllers that decide when to buy skill insurance and when to react.
arXiv:2606.17783v1 Announce Type: new Abstract: Consumer mixed reality (MR) headsets seamlessly blend virtual content into physical environments with sufficient fidelity that users may be unable to distinguish virtual objects from physical ones. We identify this virtual-physical discrimination vulnerability as an exploitable security primitive. Through speculative design workshops with 12 experts from cybersecurity and MR/HCI, we develop a taxonomy of virtual-physical confusion attacks and implement four proof-of-concept attacks on Apple Vision Pro, evaluating them with 26 participants in realistic MR tasks. All four attacks altered user behavior, with success rates ranging from 85% to 100%, producing misdirected interactions, misjudged object identities, biased purchasing decisions, and altered navigation paths. Notably, the most successful attacks were also the hardest to detect according to participants' subjective ratings. Even participants who recognized virtual content still complied behaviorally, and no participant attributed anomalous events to adversarial causes. We propose platform-level provenance, interaction gating, and user education as countermeasures.
arXiv:2606.17299v1 Announce Type: new Abstract: Word2Vec's effectiveness at generating semantic embeddings has been widely validated, yet it has been tested almost exclusively on languages with large vocabulary inventories. This study examines whether Word2Vec can successfully capture semantic relationships within an extremely reduced vocabulary using data from Toki Pona, a constructed language with approximately 130 words. We sourced 1.4 million sentences (7.95 million tokens) from the Toki Pona community for training. Approximately 23% of sentences in the corpus contain non-Toki Pona tokens such as named entities, loanwords, and neologisms. To investigate whether this linguistic noise enhances or hinders performance -- a topic rarely addressed in word embedding literature -- we trained two distinct models: one retaining these incidental tokens and another filtering them out completely. Evaluation was conducted using quantitative methods measuring word proximity to semantic category centroids, automated silhouette scores via agglomerative clustering, and qualitative analysis utilizing representational similarity matrices compared against English. The results indicate that while sparse, non-core tokens do not affect the relative structure of the learned embeddings, they actually draw similar words closer together in the vector space. Importantly, Word2Vec's effectiveness depends more on distributional patterns than lexicon size even at this extreme lower bound.
arXiv:2606.17315v1 Announce Type: new Abstract: Linear probing is one of the simplest and most space-efficient approaches to hash table design, and is widely used in sequential settings due to its compact memory layout. However, designing a concurrent linear-probing hash table with strong liveness guarantees has proved difficult, and only a handful of such algorithms have been proposed, all of which either restrict concurrency or rely on large per-entry metadata, thereby compromising space efficiency. We present a lock-free linear-probing hash table with wait-free lookups that retains the core advantages of sequential linear probing while handling contention gracefully. Our design uses only a small amount of metadata per table entry: a constant number of additional bits when using LL/SC, or a logarithmic number of bits when using CAS. The algorithm is linearizable and lock-free, supports insert, delete, and wait-free lookup operations, and is able to safely reclaim space used by deleted elements without rebuilding the table. We analyze the amortized step complexity of our hash table assuming no concurrent insertions of the same key, and show that each operation has expected amortized step complexity matching that of sequential linear probing, up to the point contention per key.
arXiv:2606.17786v1 Announce Type: new Abstract: Training psychotherapists in evidence-based interventions such as Acceptance and Commitment Therapy (ACT) requires repeated practice with meaningful feedback, yet opportunities for safe, standardized training are limited by ethical, logistical, and resource constraints. We introduce a system designed to support ACT-oriented psychotherapy training through spoken dialogue with an embodied virtual patient. The system uses large language models to simulate patient behavior conditioned on profiles derived from real therapy sessions and configurable clinical scenarios, while a separate automated evaluator provides turn-by-turn feedback on therapist responses based on established ACT fidelity criteria. Rather than aiming to replace supervision, the system is intended to support deliberate practice by enabling experimentation, reflection, and immediate feedback in low-risk settings. Expert evaluation with practicing psychologists confirmed high realism in patient behavior and demonstrated that immediate turn-by-turn ACT feedback increased therapists' awareness of intervention choices and enabled effective experimentation with alternative responses. Quantitative evaluation across 49 therapy transcripts identified GPT-4o-mini as the optimal feedback model, achieving the lowest mean absolute error (MAE = 6.12) in replicating human supervisor ACT fidelity ratings with statistically significant agreement. This work demonstrates the potential of fidelity-aware simulated patients as a scalable complement to psychotherapy training.
arXiv:2606.17821v1 Announce Type: new Abstract: Large Language Models (LLMs) have demonstrated remarkable capabilities in translating natural language to SQL, yet existing methods still falter on complex queries requiring multi-step, data-aware reasoning. We introduce DecoSearch, a training-free framework that addresses this by routing each query to the appropriate level of reasoning effort. A lightweight Schema Selector first prunes the full database schema to the relevant tables and columns. An LLM Judger then decides whether the question requires decomposition: straightforward questions follow a direct generation path and complex ones are escalated to a Directed Acyclic Graph (DAG) of atomic sub-questions, each solved by a targeted SQL generation step. A RAG component grounds the decomposer with semantically similar training examples, and a Topology Refiner restructures the reasoning plan when execution failures signal a flawed decomposition rather than a fixable SQL error. DecoSearch achieves 70.53% execution accuracy on BIRD and 88.31% on Spider with a DeepSeek backbone, surpassing all training-free baselines while consuming an order of magnitude fewer tokens than competing methods. It also functions as a model-agnostic wrapper, consistently improving fine-tuned SQL generation backbones without any modification to the pipeline.
arXiv:2606.17340v1 Announce Type: new Abstract: Accurate vision-based navigation in monocular endoscopy is difficult due to limited depth cues, weak tissue texture, non-rigid deformation, and substantial appearance variation across domains, all of which complicate pose estimation, depth prediction, and image-to-anatomy alignment. Although recent vision foundation models have shown promise, their learned representations often remain insufficiently geometry-consistent, hindering stable feature correspondence and limiting their reliability for downstream navigation tasks. We propose a unified framework for learning geometry-consistent and domain-robust image representations for monocular endoscopy. The framework combines a synthetic data pipeline that provides accurate geometric supervision with Hierarchy-Aware Geometry-Semantic Adaptation, a structured alternative to standard LoRA that inserts low-rank adapters selectively across the transformer hierarchy and couples them with layer-wise training objectives to encourage geometric correspondence in intermediate features and semantic consistency in deeper features. Experiments on public and proprietary datasets show improved geometric and semantic representation quality, leading to better performance on downstream navigation tasks including pose estimation and monocular depth estimation. The learned representations show favorable synthetic-to-real transfer on clinical bronchoscopy and provide a useful initialization for adaptation to sinus endoscopy and colonoscopy under limited supervision. The framework also shows favorable scaling with model size and training data. These results support hierarchy-aware, geometry-guided adaptation as a practical approach for endoscopic representation learning.
arXiv:2606.17347v1 Announce Type: new Abstract: This article proposes a novel methodology for the classification of transient and stationary regimes in dynamic systems. Several sensor-based solutions for regime classification in the literature require the setting of several parameters, or are not suitable for scenarios involving multivariate systems that may contain periodic signals. The proposed method introduces a spatial curve representation of the considered system based on its sample mathematical moments. Then, by connecting concepts of stability theory, geometrical properties of spatial curves and stationary stochastic processes, two regime classifiers are designed using the arc length and the curvatures of the proposed curve. Both classifiers are capable of describing and detecting transient regimes, considering behaviors such as: multivariate asymptotically, marginally stability, and cyclostationarity. Furthermore, a quantitative comparison in performance and computation resources of the proposed classifiers against existing classifiers in the literature illustrates that the proposed regime classifier based on the arc length outperforms other techniques in classifying transient regimes for simulated linear, non-linear, and discontinuous multivariate systems under the specified studied conditions.
arXiv:2601.22495v2 Announce Type: replace Abstract: Fine-tuning flow matching models is a central challenge in settings with limited data, evolving distributions, or computational constraints. While recent work has produced significant advances, particularly in the area of reward-based fine-tuning, current methods fail to demonstrate both theoretical correctness as well as strong empirical results in terms of stability, efficiency, and diversity preservation. In this work, we propose Gradual Fine-Tuning (GFT), a simple yet principled annealing-based framework for fine-tuning flow generative models when only samples from the target distribution are available. For stochastic flows, GFT defines a temperature-controlled sequence of intermediate objectives that smoothly interpolate between the pretrained and target drifts, provably approaching the true target as the temperature approaches zero. We analytically demonstrate that sample generation after GFT can be made substantially more efficient with the use of arbitrary (e.g., optimal transport) couplings, as well as by utilizing few-step inference methods. Empirically, GFT significantly improves convergence stability, while maintaining or improving generation quality, training speed, and generation diversity compared to other fine-tuning methods. Our results position GFT as a simple yet theoretically grounded and practically effective alternative for scalable adaptation of flow matching models under distribution shift.
arXiv:2602.02881v2 Announce Type: replace Abstract: This paper articulates a long-term research vision for formal methods at the intersection with artificial intelligence, outlining multiple conceptual and technical dimensions and reporting on our ongoing work toward realising this vision. It advances a forward-looking perspective on the next generation of formal methods based on the integration of automated contract synthesis, semantic artifact reuse, and refinement-based theory. We argue that future verification systems must builds towards individual correctness proofs toward a cumulative, knowledge-driven paradigm in which specifications, contracts, and proofs are continuously synthesised and transferred across systems. To support this shift, we outline a hybrid framework combining large language models with graph-based representations to enable scalable semantic matching and principled reuse of verification artifacts. Learning-based components provide semantic guidance across heterogeneous notations and abstraction levels, while symbolic matching ensures formal soundness. Grounded in compositional reasoning, this vision points toward verification ecosystems that evolve systematically, leveraging past verification efforts to accelerate future assurance.
arXiv:2606.17810v1 Announce Type: new Abstract: In this paper, we establish a set of theoretical impossibility results, termed the No-Free-Fairness theorems, that identify three fundamental sources of disparity in learning systems. First, we show that when a task exhibits irreducible cost on a subgroup, any decision rule must trade off overall performance with disparity, yielding an inherent fairness--cost frontier. Second, we prove that even in ideal, noise-free settings where a perfectly fair and accurate solution exists, finite-sample learning alone induces nontrivial subgroup disparity, ruling out distribution-free fairness guarantees. More seriously, enforcing strict relative fairness creates a statistical bottleneck: achieving low cost may require exponentially many samples. Third, we show that limitations of the model class can independently induce disparity: if the model cannot represent accurate solutions for a subgroup, fairness remains unattainable regardless of data or training procedure. Overall, these results demonstrate that unfairness is not solely a consequence of biased data or suboptimal optimization, but arises from the intrinsic structure of decision problems, the constraints of finite data, and the expressivity of models. Our framework applies broadly beyond standard supervised learning, and suggests that achieving fairness requires explicit trade-offs and should be treated as a core design consideration.
arXiv:2504.11837v3 Announce Type: replace Abstract: Emotional support conversation (ESC) aims to alleviate people's emotional distress through effective conversations. Although large language models (LLMs) have made remarkable progress in ESC, most of these studies may not define the diagram from a state-model perspective, thereby providing a suboptimal solution for long-term satisfaction. To address such an issue, we leverage the Finite State Machine (FSM) on LLMs, and propose a framework called EmoFSM. Our framework allows a single LLM to bootstrap the planning during ESC, and self-reason the seeker's emotion, support strategy, and the final response upon each conversation turn. Substantial experiments in ESC datasets suggest that EmoFSM outperforms many baselines, including direct inference, self-fine, chain of thought, finetuning, and externally supported methods, even those with many more parameters.
arXiv:2606.17847v1 Announce Type: new Abstract: WallGo is a recently introduced strategic board game popularized by the 2025 Netflix series The Devil's Plan. Although played on a small 7 x 7 board, its combination of stone movement and wall placement yields high game-tree complexity and intricate strategic interactions. Despite its growing popularity, WallGo remains underexplored. This paper presents WallZero, an AlphaZero-based agent for the two-player WallGo setting. We introduce tailored action and feature designs to improve playing performance significantly. In the evaluation, WallZero defeats two professional Go players who participated in this study, securing on average 1.98x more territory per game. Beyond its strength, we use WallZero to assess game fairness and identify key strategies for mastering WallGo. Interestingly, our results show that the opening used in the Netflix series yields a more balanced game. Our code is available at https://rlg.iis.sinica.edu.tw/papers/wallzero.
arXiv:2606.17449v1 Announce Type: new Abstract: While Multimodal Retrieval-Augmented Generation (M-RAG) enhances Large Vision-Language Models, it remains highly susceptible to cross-modal hallucinations, causal fabrications, and sycophancy. Furthermore, existing mitigation pipelines often face an intervention paradox: static rules tend to unnecessarily disrupt accurate generations, whereas leaving the multi-modal reasoning completely unguided allows existing mismatches to cascade into severe logical fabrications. To quantify and mitigate these hallucinations, we propose a Multi-Agent system, MODE-RAG, driven by Variational Free Energy (VFE) and internal attention states to dynamically gate interventions. High-risk queries are routed to five stage-specific agents, integrating Monte Carlo Tree Search (MCTS) for rigorous causal derivation and logit perturbations to penalize sycophancy. Dedicated Correction and Overseer agents ensure formatting stability and perform post-hoc factual verification. To objectively evaluate our approach, we introduce ModeVent, a challenging subset derived from the MultiVent dataset. Extensive experiments indicate that our system effectively reduces hallucination rates and logical fabrication, significantly improving the robustness of M-RAG systems.
arXiv:2606.17465v1 Announce Type: new Abstract: We introduce Perron--Frobenius Operator Matching (PFOM), a generative framework that matches density evolution via the integral PF operator, subsuming flow, diffusion, and jump models. We prove that among Bregman divergences, only Kullback--Leibler divergence preserves equality between density-level and sample-conditioned objectives, yielding a practical loss equivalent to Koopman path matching. We further develop Nesterov-accelerated training and sampling that stabilize discretization and accelerate convergence. %On Gaussian mixtures and two-moons, PFOM achieves faster KL/$W_2$/MMD decrease and improved wall-clock efficiency with empirical validation. PFOM unifies operator-theoretic identification with modern generative modeling and opens paths to adaptive dictionaries and high-dimensional applications.