arXiv:2606.24025v1 Announce Type: new Abstract: Diffusion models have achieved strong performance in image, text-to-image, and video generation, where conditional generation is often controlled by classifier-free guidance (CFG). CFG improves condition consistency by increasing a guidance weight, but stronger guidance typically reduces diversity and distributional coverage. It remains unclear how this consistency-coverage trade-off should be controlled across the reverse trajectory, since the distribution induced by CFG is not simply the fixed-time tilted distribution given by the guided score field. To address this issue, we propose an information-theoretic framework for CFG schedule optimization. Our approach uses a clean endpoint reference to specify the desired consistency-coverage trade-off, while optimizing the actual distribution induced by the guided sampler toward this reference. We derive trajectory-level formulas to estimate the objective from samples and score evaluations, avoiding explicit density estimation. On ImageNet-512 with EDM-XXL and COCO with SD-XL, the learned schedules achieve competitive or improved trade-offs over constant guidance and allocate guidance selectively across noise levels.
Science Journals
arXiv:2606.24054v1 Announce Type: new Abstract: This paper studies algorithms for computing the minimal monomial basis for multivariate Birkhoff interpolation problems. Our approach is built around the notion of a reverse reduced set, which serves as the key tool for bridging the interpolation conditions to the monomial basis, thereby avoiding the construction and evaluation of Vandermonde matrices required in existing algorithms. For the single-node case, we prove that after Gaussian elimination on the incidence matrix, the least monomials of the polynomial set corresponding to the interpolation conditions precisely constitute the minimal monomial basis. To handle the multi-node case, we exploit the one-to-one correspondence between interpolation functionals and formal power series, whereby interpolation conditions at arbitrary nonzero nodes can all be converted to the origin. This provides both a coherent theoretical framework and a constructive algorithm for determining a proper minimal monomial basis for the general multivariate Birkhoff interpolation problem. Numerical examples demonstrate the effectiveness of the proposed algorithm.
arXiv:2606.24099v1 Announce Type: new Abstract: Algorithms have become central to scientific research in the era of artificial intelligence (AI). Although algorithm mentions in papers are often used to indicate popularity and influence, existing studies usually evaluate individual algorithms in isolation and pay limited attention to the collective influence formed through their interconnections. This study constructs large-scale algorithm co-occurrence networks in natural language processing (NLP) based on the full text of academic papers and investigates algorithm influence from a network perspective. Using deep learning models, we extract algorithm entities and build overall, cumulative, and annual co-occurrence networks. We analyze their structural characteristics and apply multiple centrality measures to assess the group influence of algorithms across the whole field and over time. The results show that algorithm networks display typical features of complex networks, with increasingly dense connections developing over approximately two decades. Classic, high-performing algorithms and those located at the intersections of different research periods tend to have high popularity, control, centrality, and balanced influence. When the influence of an algorithm declines, it usually loses its core network position first, followed by weaker associations with other algorithms. This study is the first large-scale analysis of algorithm co-occurrence networks. Covering more than four decades of academic publications, it provides a temporal and structural view of algorithm influence and offers a foundation for future research on networks linking algorithms, scholars, and tasks.
arXiv:2606.24101v1 Announce Type: new Abstract: Conventional visual navigation policies often struggle with myopic decision-making and mode collapse in complex environments. While world models offer a promising alternative, existing paradigms typically isolate perception, generation, and control, failing to capture their shared spatio-temporal dynamics. In this paper, we propose NavWM, a unified navigation world model that seamlessly integrates latent world reasoning, multimodal action prediction, and controllable visual generation. At its core, NavWM leverages latent world tokens to distill geometric and semantic priors, endowing the agent with robust structural understanding. To overcome the limitations of deterministic policies, we introduce an anchor-based multimodal trajectory forecasting framework that generates a diverse action space. This inherent diversity explicitly empowers the generative world model to act as a robust closed-loop planner, utilizing visual foresight to evaluate and select the optimal path. Extensive experiments across diverse robotics datasets demonstrate that NavWM significantly advances the state-of-the-art, delivering remarkable improvements in both high-fidelity future state generation and zero-shot navigation success.
arXiv:2606.24104v1 Announce Type: new Abstract: Generative artificial intelligence(GenAI) is reshaping learning in higher education, with particularly pronounced implications for the humanities and social sciences(HSS), where learning outcomes are commonly expressed through written and interpretive forms that align closely with GenAI's capabilities. Yet, systematic evidence on the educational impacts of GenAI on HSS students remains limited. Addressing this gap, this study draws on a large-scale survey of HSS students in China to examine its role in academic development. Guided by relevant learning theories, this study focuses on four dimensions: patterns of use, effects on learning processes and academic performance, challenges associated with GenAI use, and preferred approaches to curricular integration. We found that more than half perceived enhanced learning motivation, independent thinking and creativity, although a substantial minority reported little change or even decline. Comparatively, a notably larger majority reported academic performance gains, although these gains may partly reflect limitations in conventional assessment practices. The study identifies variations in perceived learning and performance improvements among students with differing durations of GenAI experience, along with observable disciplinary differences and modest gender differences. While an overwhelming majority valued the importance of ethical considerations, only slightly more than half were satisfied with privacy protection. Limited accuracy and overreliance emerged as the most pressing concerns reported by students. Students favored partial or optional curricular integration supported by practice-oriented training, and widely recognized GenAI's significance for their future professional development. Grounded in student perspectives, this study offers evidence-based recommendations for the responsible and pedagogically meaningful integration of GenAI
arXiv:2606.24122v1 Announce Type: new Abstract: Existing head pose datasets predominantly feature subjects of Western or East Asian origin, leaving South Asian populations, particularly Bengali individuals, largely underrepresented. We introduce Bengal-HP_RU, the first publicly available head pose dataset centred on Bengali subjects, comprising 12,894 labelled head images annotated with continuous yaw, pitch, and roll values. Images were collected from Wikimedia Commons under free licences and processed through an automated pipeline followed by manual label correction. The dataset is partitioned by Wikimedia uploader identity to prevent data contamination, yielding 10,494 training and 2,400 test images across 296 unique uploaders. Bengal-HP_RU exhibits substantial diversity in subject age, gender, occlusion, illumination, and background, reflecting realistic in-the-wild conditions. The dataset is publicly available at https://doi.org/10.17632/xbw9kr37jb.2.
arXiv:2606.24145v1 Announce Type: new Abstract: Large language models (LLMs) can produce clinically fluent recommendations for type 2 diabetes while failing to satisfy guideline constraints or explicitly justify lifestyle-related glycemic claims. We present T2D-Bench, a reproducible benchmark and evidence-gated evaluation framework for testing whether LLM outputs satisfy explicit, graph-checkable evidence requirements. T2D-Bench is built on a multi-layer clinical-lifestyle knowledge graph that combines a biomedical spine (UMLS, DrugBank, SIDER), computable ADA Standards of Care rules, and lifestyle knowledge connected through a mechanistic bridge to glycemic laboratory effects. Across 100 structured vignettes spanning diagnosis, medication safety, and adversarial lifestyle conflicts, baseline outputs failed benchmark-defined evidence-path checks in 35% of cases for GPT-4o-mini and 33% for GPT-4o. The evidence gate detects unsupported omissions and uses constrained revision to bring outputs into verifier-level compliance with benchmark-defined evidence requirements. These results show that computable evidence constraints can make unsupported clinical omissions explicit, measurable, and correctable in diabetes-focused LLM outputs.
arXiv:2606.24166v1 Announce Type: new Abstract: Large Language Models remain vulnerable to adversarial prompts that elicit harmful responses, and scaling red-teaming to cover a broad range of failure modes is constrained by the cost of text generation and evaluation. We present \emph{ToxSearch-S}, a speciated extension of toxicity-focused evolutionary prompt search with incremental, embedding-driven niche maintenance, together with an MPI master-worker realization that centralizes population and species bookkeeping on rank~0 while offloading prompt evolution and evaluation to $n_w$ parallel workers. Under a common budget, ToxSearch-S attains peak toxicity competitive with both ToxSearch and RainbowPlus while following a measurably less toxic best-so-far trajectory, indicating lower cumulative search pressure. Diversity is non-uni-dimensional: RainbowPlus yields greater embedding-level spread, whereas ToxSearch-S partitions high-toxicity prompts into more localized behavioral pockets, reflected by a higher DBSCAN cluster count. MPI distribution delivers substantial wall-clock gains, approximately $1.8\times$ with two workers and $3.2\times$ with four, while leaving Best@B statistically indistinguishable from sequential execution. Four-worker runs also produce significantly larger final species cardinality and more toxicity-bearing species, without a reliable gain in global peak toxicity. These results position incremental speciation as a practical quality-diversity mechanism for AI Safety and MPI as an effective means of compressing time-to-result while preserving measured search outcomes.
arXiv:2606.24169v1 Announce Type: new Abstract: Adapting a streaming speech recognition model to a new language requires choosing between two plausible warm starts: a multilingual (ML) encoder or an English-only (EN) encoder. The common intuition is that the multilingual encoder should help most at low data, but it is unclear how long that advantage persists, whether tight streaming latency amplifies it, and whether it survives deployment quantization. We answer these questions with a controlled sweep of a 0.6 B-parameter cache-aware FastConformer transducer across eight European languages, up to five target-language data scales (100 h to 2500 h), three streaming tiers plus offline decoding, and up to four public test sets. The main result is that multilingual initialization is a data-limited advantage, not a latency-limited one. On FLEURS at 160 ms, the mean EN-ML word error rate (WER) gap falls from +4.21 percentage points (pp) at 100 h to +0.20 pp at 2500 h; a power-law fit summarizes this decay, with each doubling of target-language data roughly halving the remaining advantage. Across the three streaming tiers, the across-language mean EN-ML gap is approximately stable at each scale from 100 to 1000 h, and is near zero by 2500 h. Finally, 4-bit weight-only encoder quantization at the matched 560 ms streaming tier reduces the encoder footprint by about 3x, with an average FLEURS WER increase of about 0.5 pp. The resulting guideline is simple: use multilingual initialization in low-data regimes, treat the choice as effectively irrelevant at large data, and make latency and quantization decisions independently.
arXiv:2606.24192v1 Announce Type: new Abstract: Unlearning has emerged as a key technique to mitigate harmful content generation in diffusion models. However, existing methods often remove not only the target concept, but also benign co-occurring concepts. As illustrated in Fig.1, unlearning nudity can unintentionally suppress the concept of person, preventing a model from generating images with person. We define these undesirably suppressed co-occurring concepts that must be preserved CARE (Co-occurring Associated REtained concepts). Then, we introduce the CARE score, a general metric that directly quantifies their preservation across unlearning tasks. With this foundation, we propose ReCARE (Robust erasure for CARE), a framework that explicitly safeguards CARE while erasing only the target concept. ReCARE automatically constructs the CARE-set, a curated vocabulary of benign co-occurring tokens extracted from target images, and leverages this vocabulary during training for stable unlearning. Extensive experiments across various target concepts (Nudity, Van Gogh style, and Tench object) demonstrate that ReCARE achieves overall state-of-the-art performance in balancing robust concept erasure, overall utility, and CARE preservation.
arXiv:2606.24201v1 Announce Type: new Abstract: This paper presents FORESEE, a novel cooperative lane change model for connected and automated driving. FORESEE leverages Vehicle-to-Everything (V2X) data to anticipate traffic conditions and effectively organize lane changes. Specifically, it uses V2X data to organize vehicles into lanes based on their desired speeds, which helps to homogenize traffic flow and reduce disturbances caused by speed differences among vehicles within the same lane. The study demonstrates that implementing cooperative lane changes with FORESEE enhances average vehicle speed and energy efficiency compared to non-cooperative lane changes, which typically rely on short-term and local information about the ego vehicle and its immediate neighbors. This is achieved through fewer but more effective lane changes. Additionally, vehicles can maintain speeds closer to their desired speeds, resulting in fewer fluctuations in speed and acceleration and enhanced driving comfort. Moreover, cooperative lane changes can better manage road traffic disturbances, such as obstacles, by anticipating traffic conditions and organizing lane changes ahead. FORESEE serves as a valuable framework for the future design and testing of V2X-based maneuver coordinations as their effectiveness depends on how vehicles change lanes and their ability to plan and organize maneuvers in consideration of the upcoming traffic conditions.
arXiv:2606.24256v1 Announce Type: new Abstract: Physical interactions follow a long-tailed distribution: a set of common and regular interactions dominates human experience and visual data, while a broad spectrum of rare and irregular interactions remains underrepresented. Although recent visual world models, including image and video generation models, achieve impressive realism on existing benchmarks, they primarily focus on simulating common physical interactions. This raises a central question: Do current visual world models internalize and generalize physical principles? In this work, we introduce Tailor-Bench, a benchmark that challenges world models to simulate irregular physical interactions. To enable systematic evaluation, we design three scenario modes that progressively challenge model reasoning: Regular scenarios reflect common tool-task pairs, Unconventional scenarios replace conventional tools with attribute-compatible substitutes to test affordance generalization, and Impossible scenarios introduce attribute-violating tools to probe constraint awareness. Additionally, we design two complementary settings under a unified evaluation protocol: predictive generation requires inferring outcomes without guidance, while descriptive generation specifies the target outcome for faithful realization. Our experimental results reveal a clear long-tail gap in physical world modeling: performance degrades from Regular to Unconventional and Impossible scenarios, indicating limited generalization beyond common interactions. Failure analysis further shows that models rely on superficial visual patterns: image models fail to realize correct state changes, while video models further suffer from temporal inconsistencies.
arXiv:2606.24384v1 Announce Type: new Abstract: Ultra-high dose rate (UHDR) irradiation used in FLASH radiotherapy induces strong space-charge effects in plane-parallel ionisation chambers (PPICs), leading to significant reductions in charge collection efficiency (CCE). To investigate these effects, we extended the Garfield++ framework by implementing ion-ion recombination and self-consistent space-charge electric field calculations. The developed Monte Carlo model couples particle transport, electron attachment, recombination processes, and dynamic electric-field distortions. The implementation was validated against analytical and numerical models from the literature, including the works of Fenwick and Kumar, Kranzer et al., and Paz Mart\'in et al., with excellent agreement for the free electron fraction (FEF), CCE, induced current, and electric field evolution. The simulations show that space charge can locally increase the electric field by more than a factor of four or reduce it to nearly zero. The results suggest that CCE reduction under UHDR conditions is mainly driven by the decrease of the FEF caused by electric-field-dependent electron attachment, indicating that recombination may be largely governed by FEF evolution. This opens promising perspectives for improved analytical models and real-time correction methods for ionisation chamber dosimetry under UHDR conditions.
arXiv:2606.24491v1 Announce Type: new Abstract: In earlier work \cite{sharma2025} we developed deterministic analytical bounds on the T-eigenvalues of symmetric positive definite (SPD) third-order tensors under the Kilmer--Martin T-product: the trace--determinant (TDet) bounds via the AM--GM inequality, and the trace-dependent (TDep) bounds generalizing Samuelson's inequality. While these bounds are cheap and guaranteed-valid, their relative gap grows as $\sqrt{d-1}$ in the tensor dimension $d = np$, limiting their usefulness for large tensors. This paper develops randomized estimators for the extreme T-eigenvalues of T-SPD tensors that complement the deterministic bounds. We adapt the Halko--Martinsson--Tropp framework \cite{halko2011} to the T-product setting and introduce four methods: (i) a randomized power method that produces a lower bound on $\lambda_1$ with exponential convergence; (ii) a randomized subspace iteration with a tensor-analogue HMT error bound; (iii) a two-sided rigorous bracket combining the randomized lower bound with the deterministic TDep upper bound; and (iv) a Hutchinson-based fully randomized TDep bound for matvec-only settings.
arXiv:2606.24499v1 Announce Type: new Abstract: Existing automotive event datasets rely on appearance-based annotations from frame pipelines, making them poorly suited for motion-aware event perception. We present a geometry-driven, annotation-free framework that classifies detected objects as static or independently moving by exploiting ego-motion structure directly from the event stream. A Focus of Expansion model with yaw compensation estimates global background motion, while objects are labeled as moving when local motion deviates from this prediction, as quantified by a scale-invariant residual. Temporal stabilization improves robustness across consecutive event windows. The method requires no learning, no manual motion labels, and works with any input bounding boxes. Experiments on MVSEC and the Prophesee 1 Megapixel Automotive Detection dataset demonstrate consistent performance across diverse driving scenarios, with yaw compensation improving results during turns and a simple translational local model offering a favorable accuracy-efficiency trade-off.
arXiv:2606.24526v1 Announce Type: new Abstract: Large language models are increasingly deployed as agents that reason over documents rather than answer from parametric knowledge. We study archive-grounded reasoning: locating sparse evidence across a large, messy collection of workplace files, reconciling inconsistent terminology, units, and time conventions, and computing an answer. Existing benchmarks address only parts of this setting and none jointly stresses archive-groundedness, agentic exploration, and cross-domain coverage. We introduce Agora, a benchmark pairing 362 questions with eight domain collections of 9,664 authentic documents and 372M tokens, far exceeding any model's context window, so agents must explore deliberately rather than scan exhaustively. Agora is built by an agentic pipeline combining cross-document task synthesis, leakage-preventing obfuscation, and difficulty filtering. Evaluating eight models, we find the task far from solved: even the strongest reaches only 59.4% accuracy, with notable variation across domains.
arXiv:2606.24549v1 Announce Type: new Abstract: Full-system rehosting plays a critical role in the security analysis of Linux-based firmware. It matches commonly deployed firmware with sufficient background knowledge. However, for custom devices, existing approaches struggle to handle initialization and runtime obstacles in the rehosting process caused by specialized architectures and hardware-dependent configuration, which heavily rely on expert intervention. This ultimately creates fundamental bottlenecks and results in low rehosting efficiency. To address the above challenges, we propose FirmCure, the first LLM-driven full-system rehosting framework designed for autonomous and adaptive rehosting of Linux-based firmware. FirmCure develops an Adaptive Perception Inference mechanism to extract firmware structural dependencies via static analysis, followed by a Reflective Synthesis module for iterative configuration optimization, and finally an Autonomous Runtime Intervention module for real-time error remediation through runtime fault diagnosis and monitoring. We evaluated 21 IoT firmware images from 10 vendors across 5 architectures, while FirmCure achieved a 100% network port opening rate and 90.5% service interactivity, substantially outperforming state-of-the-art baselines. Our experiments confirm that FirmCure's intervention strategies generalize across heterogeneous firmware. The framework successfully reproduces known vulnerabilities and discovers new security flaws.
arXiv:2606.24551v1 Announce Type: new Abstract: Computer-use agents can execute software tasks through either graphical interfaces or programmatic command interfaces, but existing evaluations confound interaction modality with differences in tasks, initial states, verifiers, and permitted actions. We introduce a matched execution-layer benchmark of 440 desktop tasks across 18 applications and 12 workflow categories, where screen-only GUI agents and skill-mediated CLI agents receive identical goals, states, and final-state verifiers while being restricted to modality-native actions. In this controlled setting, the strongest GUI agent reaches a 59.1% full pass rate, outperforming the strongest original-skill CLI agent at 48.2%; however, verifier-guided skill augmentation raises CLI success to 69.3%, showing that much of the CLI deficit comes from incomplete skill coverage rather than model capability alone. These results suggest that GUI and CLI expose different execution bottlenecks: GUI agents are limited by reliable grounded interaction over long-horizon workflows, whereas CLI agents are limited by the coverage and scalability of their skill interfaces.
arXiv:2606.24585v1 Announce Type: new Abstract: While the validity of LLMs' use in the legal context remains subject to ethical and legal debate, legal professionals are already experimenting with personal LLMs, if only for translation and reformulation. However, even such a seemingly innocuous use can introduce biases through case processing speed if LLM assistants selectively refuse assistance on certain topics. To better anticipate such biases, we investigate several modern small LLMs that are most likely to be used as on-device assistants, to assess the impact of overrefusal on legal prompts. Surprisingly, we find that authority-style prefixes (``you are acting as an assistant of the national supreme court'', ``[...] defense lawyer'') systematically increase refusal rates by 2--20x over the no-prefix baseline, while a known role-play jailbreak prefix shows mixed effects, sharply increasing refusals in some models and barely shifting them in others. The finding suggests that small on-prem deployable LLMs are unstable under contextual framings that a real institutional user might naturally introduce, and further investigation is essential to minimize opportunities for bias.
arXiv:2606.24589v1 Announce Type: new Abstract: Scaling adversarial evaluation of large language models requires both a method for generating hard inputs and a reliable way to confirm that resulting failures are real. We present AdversaBench, an end-to-end red-teaming pipeline that mutates seed prompts with five structured operators, queries a target model, and confirms failures through a three-judge panel with a meta-judge tiebreaker. We report experiments on 45 seeds across three categories: reasoning, instruction-following, and tool use. Every seed produced a confirmed failure. Four findings stand out. First, operator effectiveness varies sharply by category: inject_distractor scores 0.00 mean reward on instruction-following seeds but 0.80-0.83 on reasoning and tool-use. Second, binary failure rate hides difficulty: instruction-following seeds required 2.4 attacker iterations on average versus 1.1 for other categories, a gap visible in survival curves. Third, pairwise judge agreement of 80-87% coexists with near-zero Cohen's kappa due to label skew; category-level disagreement rates are more informative. Fourth, adversarial prompts generated against Llama 3.1 8B transfer zero-shot to Llama 3.3 70B, suggesting the mutations exploit general behavioral patterns rather than model-specific weaknesses. Code, dataset, and analysis scripts are available at https://github.com/khanak0509/AdversaBench .
arXiv:2606.24597v1 Announce Type: new Abstract: A world model predicts environment dynamics based on current observations and actions, serving as a core cognitive mechanism for reasoning and planning. In this work, we investigate how world modeling based on language models can further push the boundaries of general agents. (i) We first focus on building foundation models for agentic environment simulation. We introduce Qwen-AgentWorld-35B-A3B and Qwen-AgentWorld-397B-A17B, the first language world models capable of simulating agentic environments covering 7 domains via long chain-of-thought reasoning. Leveraging more than 10M environment interaction trajectories of 7 domains in real-world environments, we develop Qwen-AgentWorld through a three-stage training pipeline: CPT injects general-purpose world modeling capabilities from the state transition dynamics and augmented professional corpora, SFT activates next-state-prediction reasoning, and RL sharpens simulation fidelity through a tailored framework with hybrid rubric-and-rule rewards. To evaluate language world models, we present AgentWorldBench, a comprehensive benchmark constructed from real-world interactions of 5 frontier models on 9 established benchmarks. Empirical results demonstrate that Qwen-AgentWorld significantly outperforms existing frontier models. (ii) Beyond foundation models, we further investigate two complementary paradigms through which world modeling enhances general agents. First, as a decoupled environment simulator, Qwen-AgentWorld supports scalable and controllable simulation of thousands of real-world environments for agentic RL, yielding gains that surpass real-environment training alone. Second, as a unified agent foundation model, world-model training acts as a highly effective warm-up that improves downstream performance across 7 agentic benchmarks. Code: https://github.com/QwenLM/Qwen-AgentWorld
arXiv:2606.24601v1 Announce Type: new Abstract: Multi-agent reinforcement learning (MARL) addresses the problem of training multiple agents that pursue collaborative, competitive, or mixed objectives. Prior work has investigated transfer learning between source and target domains in MARL; however, the majority of existing approaches impose the constraint that the dimensionalities of the observation space and the global state space must be identical across domains. In this paper, we introduce a method that explicitly accommodates mismatched state-space dimensionalities between source and target domains. The proposed approach, ASALT, incorporates both observation-level and state-level adapters that map the target-domain observations and global states into a shared embedding space, thereby enabling more effective transfer of knowledge across both actors and critics. These adapters can generate embeddings that support efficient strategy transfer across heterogeneous domains. Experimental results on multiple configurations in standard benchmark environments demonstrate that ASALT surpasses existing baselines in terms of sample efficiency and global return in cooperative settings, but its effectiveness depends on the degree of mismatch between source and target domains. Furthermore, our findings indicate that ASALT mitigates negative transfer, which frequently constitutes a major obstacle when transferring policies between domains with differing observation and action spaces.
arXiv:2606.24602v1 Announce Type: new Abstract: Despite remarkable progress in multimodal understanding, current MLLMs still exhibit limitations in video text understanding, particularly when semantics emerge through the integration of temporally distributed textual cues across multiple frames. This perception challenge fundamentally differs from static image text understanding, yet existing datasets fail to capture: the vast majority of questions remain answerable from single frames, inadequately reflecting real-world video text comprehension demands. To address this, we present ViTexQA, a large-scale video-text QA dataset, and FrameThinker for robust multi-frame temporal reasoning. We build ViTexQA via a quality-controlled Chain-of-Thought (CoT) annotation pipeline boosted with temporal constraints; all its QA pairs demand cross-frame text fusion to solve, enforcing true temporal reliance. FrameThinker adopts two-stage training for explicit temporal modeling: CoT-Guided Supervised Fine-Tuning (SFT) generates frame-aware reasoning chains, followed by Temporally-grounded Reinforcement Learning (RL) optimized with multi-frame coherence rewards. Evaluations show our method outperforms SOTA baselines on ViTexQA, lifting ROUGE-L by 6.3%.
arXiv:2606.24609v1 Announce Type: new Abstract: Large language models (LLMs) are proposed as natural-language interfaces to power system analysis, yet existing frameworks are validated almost exclusively on synthetic benchmarks and support only deterministic studies. We present CONDUCTOR, an LLM-orchestrated digital twin for distribution grid operations. An open-weights LLM orchestrates power system analysis and optimization solvers and, unlike prior systems, also performs uncertainty-aware studies: probabilistic security assessment, robust corrective dispatch, and flexibility-envelope and hosting-capacity characterization. We test it on the Bornholm 60 kV distribution network - a real Danish island power system - using one year of smart-meter measurements. An operator case study spans deterministic assessment, probabilistic risk quantification, and robust dispatch. Across a 68-prompt behavioral catalog scoring tool use, evidence consistency, state-mutation discipline, and refusal calibration, the orchestrator answers 98.5% of tasks correctly on the first attempt - the lone failure being a missing answer, not a wrong one. The full pipeline is released open source.
arXiv:2606.24610v1 Announce Type: new Abstract: The evaluation of cultural grounding context becomes complex when multiple cultures convey the same moral lesson. This challenge is particularly relevant to large language models (LLMs), which produce narratives across a wide range of languages and cultural contexts. However, it remains uncertain whether these models preserve culturally grounded meaning when equivalent moral lessons are conveyed through distinct cultural forms. This study introduces a multilingual evaluation narrative framework that integrates a cross-linguistic collection of 414 proverbs spanning 15 languages and uses four LLMs to generate 13k narratives. By employing semantically equivalent proverbs as culturally grounded prompts, the analysis assesses whether models preserve meaning across languages, how cross-lingual conditioning influences narrative realization, and whether different model families converge on similar interpretations. Results indicate that cross-lingual prompting largely preserves proverb-level semantic meaning while systematically redistributing agency, social positioning, and narrative structure. Additionally, strong inter-model convergence is observed in both monolingual and cross-lingual settings, suggesting that multilingual LLMs rely on shared semantic abstractions despite architectural and linguistic differences. These findings shed light on the need for more comprehensive evaluations of cultural grounding. Relying exclusively on semantic similarity in multilingual narrative assessments may overestimate cultural preservation by neglecting culturally meaningful variations in narrative expression.