arXiv:2607.09718v1 Announce Type: new
Abstract: We investigate the cosmological implications of dynamical dark energy (DDE) models within an anisotropic, spatially homogeneous Bianchi Type-V spacetime framework using a $1+3$ covariant thermodynamics approach. By implementing both constant ($w$) and time-varying ($w_0, w_a$) parameterized equations of state, we evaluate the background expansion history and track linear matter perturbations via the quasi-static approximation. We confront these scenarios with the latest cosmological datasets, including the Dark Energy Spectroscopic Instrument (DESI) DR2 Baryon Acoustic Oscillations (BAO), the Union3 and Dark Energy Survey 5-year (DESY5) Type Ia Supernovae compilations, Cosmic Chronometers (CC), and Redshift-Space Distortion (RSD) measurements. Our joint statistical analyses reveal that the introduction of spatial anisotropy coupled with DDE efficiently accommodates recent late-time measurements and provides a viable mechanism to mitigate the persistent $H_0$ and $S_8$ cosmological tensions. Model selection metrics show that while Akaike criteria strongly support the extended Bianchi Type-V scenarios across most joint data combinations, Bayesian criteria continue to favor the simpler standard $\Lambda$CDM baseline due to its lower dimensionality. Finally, we establish tight constraints on the current matter density parameter $\Omega_{m,0}$, the shear parameter $\Omega_{\sigma,0}$, and the dark energy evolution parameters, confirming that anisotropic extensions remain viable and testable frameworks for modern precision cosmology.
Science Journals
arXiv:2607.10609v1 Announce Type: new
Abstract: A compressible capillary drop differs from an incompressible one in that its radius is a dynamical degree of freedom. When the drop is sufficiently small, this spherical mode can become unstable, defining a critical radius set by the competition between surface tension and compressibility. This paper examines the robustness and physical meaning of that critical radius. For a non-relativistic viscous fluid, starting from the viscous compressible equations and the free-surface stress condition, we derive the radial dispersion relation and show that shear and bulk viscosities change the eigenvalues but not the onset radius. A thermodynamic argument identifies the same radius as the point where the energy of a uniformly compressed drop changes from locally stable to unstable, explaining why the threshold is not set by viscous dissipation. This energetic interpretation can also be applied to a special-relativistic fluid, where the non-relativistic mass-density factor is replaced by the corresponding enthalpy-density factor. We then compare the critical radius with diffuse-interface length scales in two Cahn--Hilliard free-energy models to determine whether the instability can occur within the range of validity of the sharp-interface description. In a symmetric quartic model the critical radius is much smaller than the interface thickness, so the instability is absent throughout that range. In a shallow-well model, however, the critical radius can become parametrically larger than the interface thickness, leaving a range of sharp-interface drops that are unstable. Whether such a range exists therefore depends on the diffuse-interface free-energy model.
arXiv:2607.11397v1 Announce Type: new
Abstract: Generalizable robot policies typically rely on action-labeled robot demonstrations, which are expensive to collect and difficult to scale. In contrast, large-scale human and robot videos contain rich physical interactions but often lack executable robot action labels. We present WALA, a framework for learning executable latent actions from both action-labeled demonstrations and action-free videos. WALA first pretrains a semantic-geometric latent action model from videos by modeling the evolution between current observations and sparsely sampled future observations. Instead of reconstructing raw pixels, WALA predicts future deltas in the DINOv3 feature space and dense depth space, preserving task-relevant semantic and geometric structure while reducing sensitivity to appearance details. During policy training, the pretrained encoder provides stable latent action targets, and the decoder serves as a trainable latent world model. The latent actions generated by the vision-language backbone are jointly supervised by robot action prediction, latent action target matching, and future dynamics prediction. This enables action-labeled demonstrations to provide executable control supervision, while action-free videos contribute dynamics supervision without requiring robot action annotations. Experiments show that WALA achieves strong performance on RoboTwin, sets a new state-of-the-art result on RoboCasa with 75.2% average success, and improves both policy performance and generalization in real-world manipulation tasks.
arXiv:2607.10540v1 Announce Type: cross
Abstract: We propose a two-stage estimator for structural mediation parameters that combines deep representation learning with G-estimation under the "no essential heterogeneity" (NEH) assumption. We call the method UNIT. In the first stage,TARNet estimates the heterogeneous effect of a randomized treatment on a mediator by learning a shared covariate representation across treatment arms.The resulting conditional average treatment effect (CATE) estimate provides a plug-in approximation to the heterogeneity-dependent component of the weight function entering the G-estimating equation of Zheng and Zhou (2015), which identifies the structural parameters even in the presence of unmeasured mediator-outcome confounding. We show that more accurate first-stage representation learning can yield a more informative plug-in weight and thereby improve the precision of the structural parameter estimator. In simulations with non-Gaussian covariates and nonlinear mediator effects, TARNet weights reduce the Stage-2 standard error of the mediation coefficient by a factor of $1.45$ to $1.51$ (median across replications, $n \ge 2000$) relative to the classical approach, at no cost to bias or coverage.
arXiv:2607.10151v1 Announce Type: new
Abstract: Retrieval-Augmented Generation (RAG) systems are widely adopted in question answering, yet they often fail to satisfy complex multi-constraint queries, leading to constraint violations, factual inconsistencies, or hallucinations. We present Structure-Driven RAG System for Multi-Constraint Queries(MC-RAG), a structure-driven RAG system that reformulates retrieval as a subgraph matching problem over a knowledge graph. By integrating semantic and structural embeddings with path-level indexing, MC-RAG performs interpretable, structure-aware, and constraint-consistent retrieval and generation. During the demonstration, participants can input medical or encyclopedic multi-constraint queries, visualize how the system parses constraints, performs structural matching, and generates answers, thereby experiencing an end-to-end, interactive, and explainable RAG pipeline. A demo video is available at https://youtu.be/J8kahzmAnu0.
arXiv:2607.10745v1 Announce Type: new
Abstract: This paper describes the first ChineseBabyLM challenge, which will be held in the 2026 NLPCC conference. The challenge calls for researchers to train language models from scratch with 100 million Chinese tokens and evaluates the models on 3 tracks of tasks: NLU, cognitive alignment and Hanzi knowledge. There is no restriction on tokenizer, model architecture and the number of training epochs. Details of the challenge can be found in https://chinese-babylm.github.io/.
arXiv:2607.11424v1 Announce Type: new
Abstract: Graph sampling techniques have been widely used to simplify network computation and visualization, which also results in inevitable differences between the sampled networks and the original networks in terms of nodes, edges and structures. Investigating such differences can inform graph sampling technique users of the pros and cons of different techniques and select the appropriate one, and can also help graph sampling developers evaluate their own technique. However, there are still no systematic ways to achieve such a goal. This paper fills this research gap by first proposing systematic and generic quantitative measures to quantify three categories of graph differences (i.e., neighbor-based, path-based, and structure-based). Built upon this, we further propose DiffLens, a novel visualization system to help graph sampling developers and users intuitively explore local differences at different regions of their interest within a sampled graph, where three new lens-based visual designs are presented to display the neighbor-based, path-based, and structure-based differences respectively. We conducted two case studies and a user study using real-world network datasets to evaluate DiffLens. The results confirmed its effectiveness and usability in helping users explore local differences and compare different graph sampling strategies.
arXiv:2607.11435v1 Announce Type: new
Abstract: Linearizability is a standard correctness condition for concurrent data structures. It guarantees that operations behave as if they took effect at some atomic instant between their call and return points. Despite the central role linearizability plays, prior work has argued for instead using a style of specification that internalizes the atomicity of operations in terms of the logic's reasoning rules, known as logical atomicity. These logically atomic specifications are intended to be easier to compose inside of the logic than linearizability. Prior work has shown that in the Iris separation logic framework, a certain form of logically atomic specifications implies that a data structure is linearizable. However, the converse remained an open question: for every linearizable data structure, is it always possible to derive a corresponding logically atomic specification?
This paper resolves this question in the affirmative. We prove a completeness theorem for Iris that derives a logically atomic specification for any linearizable data structure. As a consequence, we are able to embed a variety of linearizability proof techniques into Iris and use them to derive logically atomic specifications. We apply this to three linearizability proof methods: aspect-oriented linearizability proofs, forward simulations with commit points, and meta-configuration tracking. Using these embeddings, we derive logically atomic specifications for the Herlihy-Wing queue and the Baskets Queue. We furthermore establish a connection between logical atomicity and an encoding of refinement in Iris that has been used in prior logical relations models. This result allows us to transport logically atomic specifications across refinements, which we apply to the Folly MPMC queue implementation. All of the results in this paper have been mechanized in the Rocq Prover.
arXiv:2607.11445v1 Announce Type: new
Abstract: A substantial number of patients experience diminished mobility due to disabilities, diseases, or accidents. Although modern prostheses, powered by deep neural networks, hold the promise of significantly enhancing the quality of life for these individuals, their widespread adoption is hindered by significant latency, energy consumption, and spatial requirements. Wired connections to external high-performance processors restrict patient mobility, while wireless connections limit the volume of information that can be transmitted to these processors. Spiking neural networks offer the potential for compressed communication and low-power inference, yet they often lag behind state-of-the-art deep learning models in various applications. In this study, we propose a high-performance neural decoding method that effectively balances task performance and efficiency. An eventbased gated recurrent unit generates a sparse communication pattern with graded spikes, surpassing classical spiking neural networks in terms of task performance. Utilising an efficient training method and sparse inference, our model presents new opportunities for on-device neural decoding.
arXiv:2607.11459v1 Announce Type: new
Abstract: This paper presents the mAIEnergy dataset, an open-access, multimodal corpus developed to support Large Language Model (LLM) applications in the energy sector. The dataset integrates approximately 50,000 textual documents, 20,000 images, 25 million numerical time series records, and 2 million geospatial and relational data entries. It includes policy and regulatory texts, scientific articles and news articles, satellite and contextual imagery, electricity system measurements, weather observations, statistical indicators, and geospatial representations of energy infrastructure and related entities. All data have been harmonized into structured, ready-to-use formats, accompanied by consistent metadata and reproducible data retrieval and preparation workflows. The dataset can serve as a foundational energy knowledge base, allowing energy stakeholders to integrate additional open-source or proprietary data. The mAIEnergy dataset adheres to Findable, Accessible, Interoperable, and Reusable (FAIR) principles, enhancing its applicability for AI-driven energy research, modeling, and decision-making.
arXiv:2607.10554v1 Announce Type: cross
Abstract: With the development of generative AI, watermarking techniques have been widely used to detect the authenticity of AI-generated data and protect the rights of users and creators. While it is already well applied in data types including imaging and text data, watermarking tabular data is still under-explored. Existing methods primarily focus on numerical data, leaving discrete, categorical, and mixed data less studied. In this work, we propose STAMP (Single-observation Tabular Attribution and Marking Procedure), a novel framework for watermarking tabular data that can accommodate and preserve a wide range of distributions. We also develop a corresponding detection mechanism, which can reliably identify watermarks even when the sample size is as small as one. We establish theoretical guarantees for asymptotic consistency and detection accuracy. Finally, through extensive simulation studies and two real-data applications, we demonstrate that the proposed method is effective and robust to subsetting, while maintaining data fidelity and a high detection rate.
arXiv:2607.10628v1 Announce Type: new
Abstract: We present Anamnesis, an interactive system for demographically controllable survey simulation using large language models. Open-source, and designed for non-technical users/researchers, Anamnesis enables the prototyping and stress-testing of survey instruments on virtual populations rather than real human subjects. The platform operationalizes the recently introduced Anthology and Alterity frameworks, which use structured narrative backstories to condition model responses, within a unified web interface. It supports open-ended generation, probabilistic demographic resampling, and multimodal (image and audio) surveys. We evaluate the system through two case studies: (1) replicating segments of Pew Research Center's American Trends Panel (ATP) on political typology and biomedical issues and (2) emulating human preference in the New Yorker Caption Contest. In both cases, Anamnesis produces opinion distributions that more closely match real-world survey data than standard persona-prompting baselines, offering a transparent, reproducible, and open-source alternative to proprietary simulation services.
arXiv:2607.11486v1 Announce Type: new
Abstract: Chess engines have long achieved superhuman playing strength. However, the underlying strategy behind their move suggestions is difficult for human players, even skilled ones, to comprehend. Motivated by this, we propose the task of chess strategy verbalization, which is to describe chess strategies in natural language. We design (i) a pipeline for verbalizing strategies and (ii) an evaluation framework for objective evaluation of generated strategy descriptions. Our experiments show that natural language is a promising and interpretable medium for communicating strategic information to both human and LLM players. We glean additional interesting insights, including (a) the importance of evaluating strategies beyond the main line, (b) the limitations of pure concept-based descriptions, and (c) the limitations of relying on LLMs rather than humans for evaluation.
arXiv:2607.10582v1 Announce Type: new
Abstract: Large language model (LLM) agents accumulate heterogeneous context, including system instructions, plans, user turns, retrieved documents, tool outputs, and intermediate reasoning, whose key-value (KV) cache can become a major memory bottleneck. Existing eviction policies generally apply the same attention- or recency-based rule to every token, ignoring semantic structure already available to the agent orchestrator.
We introduce MemDecay, a training-free, region-aware KV-cache eviction policy. MemDecay assigns tokens region-specific base priorities and decay rates, refreshes retention scores when tokens receive attention, and evicts the lowest-scoring pages under a fixed cache budget while allowing critical regions to be pinned. We also provide a procedure for calibrating decay rates from measured attention lifetimes.
We evaluate MemDecay at approximately 450 and 1,700 token contexts using Qwen2.5-1.5B and 3B. Across all settings, attention lifetimes differ by an order of magnitude across regions: system-token half-lives range from 148 to 189 decoding steps, compared with 14 to 16 for scratchpad tokens. Pinning preserves system-region facts at full-cache accuracy in every setting, while no baseline preserves more than 13 of 24. Region-aware retention remains effective as context grows, whereas recency-based retention collapses. Accumulated-attention retention performs better on unpinned content, however, and ablations identify attention-score normalization as the main limitation of the current formulation. These results establish semantic prompt structure as a robust signal for KV-cache management while clarifying how it should be combined with attention-based importance.
arXiv:2607.11303v1 Announce Type: new
Abstract: Event cameras offer microsecond-level temporal resolution and high dynamic range for low-altitude UAV perception. However, long-range UAVs often produce sparse, fragmented, and noise-contaminated event responses, where one semantic target may appear as multiple spatially separated blobs. Direct blob-level asynchronous tracking therefore suffers from duplicate trajectories and unstable identities. We propose ASUMOT, a motion-consistency-based asynchronous UAV detection and tracking framework operating directly on raw events. ASUMOT models each UAV as a set of motion-consistent event blobs. A local motion-consistency estimator triggers reliable candidates, a lightweight multi-task verifier provides UAV confidence and motion-direction cues, and motion-consistency clustering aggregates fragmented blobs into identity-consistent UAV tracks. We also introduce ES-UAV, a high-definition event-level UAV benchmark with dense semantic annotations. Experiments on public UAV tracking data and ES-UAV show that ASUMOT improves the accuracy--efficiency trade-off while preserving asynchronous event processing. Code and Dataset will be released.
arXiv:2607.10751v1 Announce Type: new
Abstract: Following Alon, Hanneke, Holzman, and Moran (FOCS 2021), we define a partial concept class (PCC) as a family of partial functions \(f: V\to\{0,1,\ast\}\); equivalently, its concepts partition the ground set into black ($f^{-1}(1)$), grey ($f^{-1}(\ast)$), and white parts ($f^{-1}(0)$). Its VC dimension is defined by shattering sets on which the value $\ast$ is not taken. We study two geometric PCCs in real Banach spaces, both with a margin \(\delta>0\): expanded half-spaces, where the grey part is a strip of width at least \(\delta\) adjacent to a half-space, and expanded balls, where the grey part is an annulus of width \(\delta\) around a unit radius ball.
Our main results are dimension-free upper bounds on the VC dimension of the PCC of expanded balls in \(L_p\parenth{\mu}\), \(1\le p<\infty\), including the non-Euclidean and algorithmically particularly relevant case \(\ell^d_1\). These bounds depend on the margin and on the radii, but not on the ambient dimension or the underlying measure space. These are extensions of the work of Bourneuf, Charbit, and Thomass\'e (FOCS 2025) who studied the PCC of expanded balls in Euclidean space, that is, $\ell_2^d$. We also prove lower bounds on the VC dimension that match the upper bounds in terms of the margin parameter $\delta$. Finally, we derive a Dense Neighborhood Lemma in \(L_p\)-spaces, again extending the known Euclidean results.
Our method relies on the linearization of the distance through a map into a space of non-trivial Rademacher type, and then the use of a balanced signed-sum estimate, or a no-dimensional Radon theorem. The arguments rely on ideas from functional analysis that are clearly explained for the non-expert in that field.
arXiv:2607.10625v1 Announce Type: new
Abstract: Language-conditioned Imitation Learning (IL) is essential for enabling robots to perform complex tasks following natural language instructions. However, generalizing to multi-step compositional tasks remains a significant challenge. While hierarchical approaches attempt to address this by decomposing tasks into atomic skills, existing methods often suffer from training instability and codebook collapse due to the tight coupling between high-level skill reasoning and low-level action generation in joint training paradigms. Inspired by the Dual-Process Theory of cognition, we propose Dual-Process Atomic Skill Learning (DASL), a novel asynchronous hierarchical imitation learning framework that decouples slow semantic reasoning from fast, real-time motion control. DASL comprises a Slow-Frequency Policy that predicts interpretable, discrete skills via Vector Quantization, and a High-Frequency Policy that leverages a latent diffusion model and a Decision Transformer to generate precise actions conditioned on these latent skills. By asynchronously coordinating these modules and utilizing diffusion to structure the latent space, our framework mitigates the skill codebook interference problem common in joint training paradigms. Evaluations across simulation benchmarks and experiment demonstrate that DASL significantly outperforms state-of-the-art baselines, excelling in skill acquisition and compositional generalization to unseen instructions. GitHub page: https://github.com/Hatakekaka/DASL
Compound Interference Recognition for LR-FHSS Satellite IoT Uplinks via Multi-Domain Instance Fusion
arXiv:2607.11488v1 Announce Type: new
Abstract: Long range-frequency hopping spread spectrum (LR-FHSS) is a promising uplink physical layer for massive low Earth orbit satellite Internet of Things, where low power terminals report short packets from wide area regions with limited terrestrial infrastructure. However, satellite IoT links are exposed to external interference, and the coexistence of multiple interference components can severely degrade receiver reliability and complicate interference mitigation. Existing recognition methods either focus on single interference scenarios or treat each compound interference combination as an independent class, leading to limited generalization or poor scalability. To address this problem, this paper formulates LR-FHSS uplink compound interference recognition as a multi-instance multi-label learning problem and proposes a multi-domain instance fusion method. The proposed method fuses local instances from the time-frequency and frequency domains and aggregates their predictions for bag-level multi-label recognition. A dataset construction pipeline is developed based on the US915 LR-FHSS configuration and incorporates shadowed-Rician fading and time-varying Doppler to emulate practical satellite communication conditions. Considering the difficulty of obtaining labeled compound interference samples in practice, single-to-compound generalization and few-shot compound interference adaptation are investigated as two practical receiver deployment scenarios. Experimental results show that the proposed method improves the overall exact accuracy over the strongest baseline by 14.71 percentage points in single-to-compound generalization and by 14.81 percentage points in few-shot compound interference adaptation for $r=1$.
arXiv:2607.10059v1 Announce Type: new
Abstract: Agent systems based on large language models (LLMs) are increasingly deployed for autonomous tasks, yet existing evaluations mostly focus on task success rather than whether agents know when to abstain. This gap poses real risks: under ambiguity, conflicting constraints, or tool failures, agents may execute unintended and irreversible actions. To close this gap, we present the first systematic evaluation framework for agentic abstention: the calibrated ability of tool-using LLM agents to recognize when not to act. At its core, AgentAbstain is a paired-task benchmark built on an agent-native taxonomy of 8 abstention scenarios across pre-execution reasoning and runtime discovery. It contains 263 paired tasks across 42 executable sandbox environments, where each pair consists of a should-act task and a should-abstain variant produced through a controlled perturbation to the instruction, tool, or environment state. To scale this paired design and resist data contamination, we propose AbstainGen, a fully automated pipeline that synthesizes sandbox environments and generates paired tasks end-to-end, validated by deterministic replay and semantic LLM judges; fresh task instances can be regenerated on demand, and three independent annotators rate 94-98% of sampled tasks as well-designed. Across 17 frontier LLMs in 4 agent harnesses, the best agent (Gemini 3.1 Pro) achieves only 59.5% paired accuracy (correct on both the act and abstain sides of each paired task). More importantly, abstention capability is largely independent of general task-solving capability, indicating that scaling task-solving alone will not close this gap. We further identify failure modes such as post-hoc abstention, in which agents execute irreversible actions before recognizing abstention triggers. Our code and dataset are open-sourced at agentabstain.github.io.
arXiv:2607.10635v1 Announce Type: new
Abstract: A novel variant of the Jacobi-Davidson (JD) type method for Hermitian eigenvalue problems, designated as JD-V, is proposed based on a newly designed correction equation, whose solution is shown to be nearly as effective as that of the standard correction equation for subspace expansion. Rigorous convergence analysis of MINRES for solving these equations reveals that the inner iterations of JD-V are significantly more efficient than those of the standard JD method when highly clustered eigenvalues are of interest. A thick-restart JD-V algorithm with deflation and purgation is developed to compute several eigenpairs of a a large-scale Hermitian matrix. Numerical experiments confirm the theoretical results and demonstrate the considerable superiority of JD-V over standard JD in overall efficiency.
arXiv:2607.09766v1 Announce Type: new
Abstract: AI agents are increasingly deployed in shared environments where they pursue diverse goals and compete for rewards. This multi-agent competition can lead to behaviors that serve individual gains at collective cost -- for instance, marketing agents may post misleading content as a result of competing for engagement on social media. Human societies address such problems through norms that constrain acceptable behavior, supported by enforcement mechanisms that detect and penalize violations. Motivated by this, we study norm enforcement mechanisms for language model agents. We find that simple enforcement mechanisms are exploited by misaligned agents for competitive advantage, even when they are not explicitly trained or prompted to do so. We thus turn our attention to designing more robust mechanisms, and identify two key ingredients: estimating each agent's reliability over time, and updating this estimate with escalating penalties for repeated misbehavior. Across three simulated environments and a variety of agent populations, mechanisms built on these principles resist exploitation, while still penalizing norm violations at comparable or lower cost than baselines. Our results position norm enforcement mechanisms as scalable levers for shaping agents' behavior, but only when designed to anticipate becoming part of the system they govern. Our code and data are available at https://yaowenye.com/norm-enforcement.
arXiv:2607.10072v1 Announce Type: new
Abstract: This paper develops a geometry-consistent fully multiplicative unscented Kalman filter (FM-UKF) for joint spacecraft attitude--orbit estimation with simultaneous dual star-tracker misalignment calibration. The estimator uses a 21-dimensional local error state combining attitude, angular velocity, gyroscope bias, inertial position and velocity, and two tracker-misalignment vectors on a mixed quaternion--Euclidean manifold. Gyroscope, star-tracker, and planet line-of-sight measurements are fused, with celestial aberration retained to capture velocity-dependent optical coupling. A multiplicative extended Kalman filter (MEKF) is implemented as a first-order baseline using the same nominal state, attitude retraction, and unit-vector measurement geometry. Monte Carlo results show similar short-step performance, while at coarse propagation intervals the proposed FM-UKF remains consistent and the MEKF exhibits divergence.
arXiv:2607.09787v1 Announce Type: new
Abstract: LiDAR semantic segmentation is a key perception task in autonomous driving, where false predictions can affect downstream planning and safety-critical decision-making. Although adversarial attacks, and specifically adversarial examples, have been widely studied for image classification and 3D point cloud segmentation, unrestricted adversarial examples remain largely unexplored in the space of 2D range images, which are projections of 3D point clouds. The proposed method is, to the best of our knowledge, the first diffusion-based unrestricted adversarial attack against 2D range-image segmentation, using adversarial guidance from a segmentation loss. By applying guidance directly during sampling, the method produces unrestricted adversarial examples that remain close to the learned LiDAR data manifold while inducing structured segmentation errors. Experiments on the SemanticKITTI dataset using RangeNet++ and CENet segmentation networks demonstrate that the attack provides adjustable degradation across guidance strengths and transfers across segmentation architectures. Compared with norm-bounded FGSM and SegPGD baselines, the proposed attack offers a distinct effectiveness-realism trade-off, achieving controllable white-box and transfer degradation while maintaining competitive distributional and visual realism.
arXiv:2607.11503v1 Announce Type: new
Abstract: Long-form article generation remains difficult for large language models because it combines long context, long instructions, and long outputs. Existing multi-agent pipelines such as STORM improve information coverage by simulating role-specialized agents, but their capabilities are often entangled in prompts and fixed procedures, making them hard to inspect, reuse, or iteratively improve. This paper presents GEIS (Generation-Evaluation-Improvement loop of agent Skills), a loop of named and declarative skills for Wikipedia-style long-form article generation. Implemented and evaluated in Tasi Harness, GEIS composes skills for article writing, browser-based evidence and image collection, diagram rendering, PDF-aware pairwise evaluation, and rule-level skill improvement. Its core writing skill follows Request, Plan, Draft, Audit, Refine, and Deliver; the pairwise evaluation skill produces structured quality reports; and the improvement skill maps recurrent findings into permanent patches to the writing skill in our 20-topic experiment. We evaluate GEIS on 20 Wikipedia Featured Article topics. Under the same generation backend, GEIS improves over the Tasi Harness default writer by 8.0 points on a 100-point PDF quality rubric and outperforms STORM on the two comparable writing dimensions, structural quality and content quality. In the 20-topic improvement experiment, the patched writing skill raises the average score from 82.90 to 86.95, with 17 out of 20 topics improved and the gain mainly coming from content quality. These results show that long-form generation can be reframed from a fixed workflow into an inspectable, modular, and evaluation-guided improvement loop.
arXiv:2607.11515v1 Announce Type: new
Abstract: Dzongkha, being the national language of Bhutan, is a common and widely spoken language in the country. Official documents, scriptures and other literature products are written in Dzongkha in order to retain the cultural value. However, documenting Dzongkha writing is a challenging and time-consuming process, largely due to the complexity of the script, the need for multiple keystrokes per syllable, and the limited availability of efficient typing tools. An immediate system that can predict and display a list of probable words for Dzongkha is the solution for this problem. The project is mainly aimed to make Dzongkha typing as convenient as possible by reducing the number of keystrokes. Our dataset is acquired from DCDD and has a total of 100000 sentences, 1331282 words and 28344 unique words. The data preprocessing was done by removing all the alphanumeric characters, tokenization, generating N-gram sequences and padding. Three models selected for training are LSTM, Bi-LSTM and GRU. The training process included fine-tuning of the model's hyperparameters. GRU being lightweight and able to handle larger datasets performed best with 74.03% accuracy and also solved the problem of overfitting.