arXiv:2605.12824v2 Announce Type: replace
Abstract: Large language models (LLMs) can generate high-level diverse phenomena without explicitly programmed rules. This capability has led to their adoption within different agent-based models (ABMs) and social simulations. Recent studies investigate their ability to generate different phenomena of interest, for example, human behavior on social media platforms or alien behavior in game-theoretic scenarios. However, capability, prediction, and explanation are different--drawing from the philosophy of science and mechanisms literature, explanation requires showing, to some degree, how a phenomenon is produced by related organized entities and activities. For modelers, describing the characteristics of an experiment or whether a simulation provides progress in capability (or explanation), can be difficult without being grounded in potentially distant research areas. We integrate recent work on LLM-ABMs with contemporary philosophy of science literature and use it to operationalize a definition of 'plausibility' in a four-level scale. Our scale separates the evaluation of a model's generative sufficiency (ability to reproduce a phenomenon) from its mechanistic plausibility (how the phenomenon could be produced), and clarifies the distinct roles of different models, such as predictive and explanatory ones. We introduce this as the Mechanism Plausibility Scale.
Science Journals
arXiv:2605.11710v2 Announce Type: replace
Abstract: Object-centric representations promise a key property for few-shot learning: Rather than treating a scene as a single unit, a model can decompose it into individual object-level parts that can be matched and compared across different concepts. In practice, this potential is rarely realized. Continual learners either collapse scenes into global embeddings, or train with part-level matching objectives that tie representations too closely to seen patterns, leaving them unable to generalize to truly novel concepts. In this paper, we identify this fundamental structural conflict and pioneer a new paradigm that strictly decouples representation learning from compositional inference. Leveraging the inherent patch-level semantic geometry of self-supervised Vision Transformers (ViTs), our framework employs a dual-phase strategy. During training, slot representations are optimized entirely toward holistic class identity, preserving highly generalizable, object-level geometries. At inference, preserved slots are dynamically composed to match novel scenes. We demonstrate that this paradigm offers dual structural benefits: The frozen backbone naturally prevents representation drift, while our lightweight, holistic optimization preserves the features' capacity for novel-concept transfer. Extensive experiments validate this approach, achieving state-of-the-art unseen-concept generalization and minimal forgetting across standard continual learning benchmarks.
arXiv:2506.09702v2 Announce Type: replace
Abstract: Mapping National Vulnerability Database (NVD) records to vulnerability-fixing commits (VFCs) is crucial for vulnerability analysis but challenging due to sparse explicit links in NVD references. This study explores this mapping's feasibility through an empirical approach. Manual analysis of NVD references showed Git references enable over 86% success, while non-Git references achieve under 14%. Using these findings, we built an automated pipeline extracting 31,942 VFCs from 20,360 NVD records (8.7% of 235,341) with 87% precision, mainly from Git references. To fill gaps, we mined six external security databases, yielding 29,254 VFCs for 18,985 records (8.1%) at 88.4% precision, and GitHub repositories, adding 3,686 VFCs for 2,795 records (1.2%) at 73% precision. Combining these, we mapped 26,710 unique records (11.3% coverage) from 7,634 projects, with overlap between NVD and external databases, plus unique GitHub contributions. Despite success with Git references, 88.7% of records remain unmapped, highlighting the difficulty without Git links. This study offers insights for enhancing vulnerability datasets and guiding future automated security research.
arXiv:2605.08475v2 Announce Type: replace
Abstract: Mechanistic accounts of in-context learning (ICL) have identified iterative algorithms for linear regression and related linear prediction tasks, often using linear or ReLU attention variants. For nonlinear ICL, prior work has related softmax and kernelized attention to functional-gradient-type dynamics, but it remains unclear whether a standard transformer with softmax attention can implement a convergent solver with an end-to-end prediction-error guarantee. In this paper, we study in-context kernel ridge regression (KRR) with Gaussian kernels and show that a standard softmax-attention transformer can approximate the KRR predictor during its forward pass by implementing preconditioned Richardson iteration on the associated kernel linear system. Under bounded-data assumptions, we construct a single-head transformer with $O(\log(1/\epsilon))$ blocks and MLP width $O(\sqrt{N/\epsilon})$ that achieves $\epsilon$-accurate prediction for prompts of length $N$. Our construction reveals a functional decomposition within the transformer architecture: softmax attention produces a row-normalized Gaussian-kernel operator needed for cross-token interactions, while ReLU MLP layers act locally to approximate the intra-token scalar arithmetic required by the update. Empirically, we train GPT-2-style transformers on Gaussian-process regression tasks to further test the preconditioned Richardson interpretation. Through linear probing, we compare the transformer's layer-wise predictions with the step-wise outputs of classical KRR solvers and find that its error profiles align most consistently with preconditioned Richardson iteration. Ablation studies further support this interpretation. Together, our theory and experiments identify preconditioned Richardson iteration as a concrete mechanism that softmax-attention transformers can realize for nonlinear in-context Gaussian-kernel regression.
arXiv:2605.18545v1 Announce Type: new
Abstract: Fringe projection profilometry (FPP) is a widely used technique for measuring object surface form and three-dimensional (3D) geometry, capable of delivering high-precision, high-resolution measurements when paired with suitable cameras and projectors. However, in practical deployments, identifying parameter configurations that maximise precision while satisfying real-world constraints remains challenging. To address this, we present an automated digital twin framework implemented in Blender, an open-source 3D software package that provides a ray-traced rendering environment that enables accurate simulation of physical systems. We replicated the physical setup in our digital twin by matching characterisation quality, gamma response, and characterisation images. Accurate system characterisation using Zhang's method [1], to obtain intrinsic and extrinsic parameters, is shown to be critical for achieving high precision. Using this digital twin, we then demonstrate systematic exploration and optimisation of key parameters, including phase-shift count, camera-projector spacing, and fringe density. These parameters span both system geometry (e.g. camera-projector positioning) and algorithmic choices, such as 2D phase-shifting and unwrapping methods [2]. Three measurement artefacts, representative of real world metrology scenarios, were used to benchmark the system. The symmetrical mean Chamfer distance (SMCD), computed between ground-truth and reconstructed meshes, was used to evaluate reconstruction quality. After optimisation within the digital twin, transferring the optimal parameters to the physical system reduced the number of required images per measurement by 48% (from 36 to 21). A reduction of 74.0% mean SMCD was also achieved for fringe pattern stripe count alteration. A 36.9% mean SMCD was obtained for adjusting the camera and projector spacing purely in the digital-twin.
arXiv:2605.18541v1 Announce Type: new
Abstract: Modeling hyperspectral imagery (HSI) across different sensors presents a fundamental challenge due to variations in wavelength coverage, band sampling, and channel dimensionality. As a result, models trained under a fixed spectral configuration often fail to generalize to other sensors. Existing Vision Transformer (ViT) approaches either rely on implicit spectral modeling with fixed channel assumptions or adopt explicit spatial-spectral attention with prohibitive computational cost, leading to a fundamental trade-off between efficiency and expressiveness. In this work, we introduce Low-rank Efficient Spatial-Spectral ViT (LESSViT), a sensor-flexible architecture for cross-spectral generalization. LESSViT is built on LESS Attention, a structured low-rank factorization that models joint spatial-spectral interactions through separable spatial and spectral components, reducing the complexity of full spatial-spectral attention from $O(N^2 C^2)$ to $O(rNC)$, where $N$ is the number of spatial tokens, $C$ is the number of spectral channels, and $r$ is the rank of the low-rank approximation. We further incorporate channel-agnostic patch embedding and wavelength-aware positional encoding to support flexible spectral inputs. To enable efficient and robust pretraining, we introduce a hyperspectral masked autoencoder (HyperMAE) with decoupled spatial-spectral masking and hierarchical channel sampling. We evaluate LESSViT under a cross-spectral generalization setting that simulates cross-sensor variability. Experiments on the SpectralEarth benchmark demonstrate that LESSViT improves robustness under spectral shifts while remaining competitive in-distribution, and explicit and efficient spatial-spectral modeling is essential for scalable and generalizable hyperspectral representation learning.
arXiv:2505.17152v2 Announce Type: replace
Abstract: Vector search underpins modern AI applications by supporting approximate nearest neighbor (ANN) queries over high-dimensional embeddings in tasks like retrieval-augmented generation (RAG), recommendation systems, and multimodal search. Traditional ANN search indices (e.g., HNSW) are limited by memory constraints at large data scale. Disk-based indices such as DiskANN reduce memory overhead but rely on offline graph construction, resulting in costly and inefficient vector updates. The state-of-the-art clustering-based approach SPFresh offers better scalability but suffers from reduced recall due to coarse partitioning. Moreover, SPFresh employs in-place updates to maintain its index structure, limiting its efficiency in handling high-throughput insertions and deletions under dynamic workloads.
This paper presents LSM-VEC, a disk-based dynamic vector index that integrates hierarchical graph indexing with LSM-tree storage. By distributing the proximity graph across multiple LSM-tree levels, LSM-VEC supports out-of-place vector updates. It enhances search efficiency via a sampling-based probabilistic search strategy with adaptive neighbor selection, and connectivity-aware graph reordering further reduces I/O without requiring global reconstruction. Experiments on billion-scale datasets demonstrate that LSM-VEC consistently outperforms existing disk-based ANN systems. It achieves higher recall, lower query and update latency, and reduces memory footprint by over 66.2%, making it well-suited for real-world large-scale vector search with dynamic updates.
arXiv:2605.18333v1 Announce Type: cross
Abstract: Accurate and efficient time-series forecasting remains a challenging problem for both classical and quantum neural architectures, particularly in multivariate environmental settings. This work adapts the Quantum Leaky Integrate-and-Fire (QLIF) spiking neural network for time-series regression tasks, specifically short-term multivariate weather forecasting. We extend QLIF beyond classification and demonstrate its applicability to continuous-valued prediction problems.
The QLIF-CAST model encodes neuron excitation states as single-qubit quantum superpositions, driven by Rx rotation gates and T1 relaxation decay, and is embedded within a hybrid quantum-classical recurrent architecture. We conduct two distinct evaluations. First, a controlled comparison against a parameter-matched classical LIF baseline on a multivariate weather dataset shows that QLIF-CAST achieves 15.4% lower MSE and 4.4% lower MAE, demonstrating that quantum neuronal dynamics reduce prediction error over classical equivalents. Second, a cross-domain comparative analysis with state-of-the-art quantum LSTM (QLSTM) and quantum neural network (QNN) models on air quality and wind speed benchmarks reveals that QLIF-CAST converges in up to 94% less training time, occupying a distinct position in the speed-error trade-off space. Hardware verification on IBM Marrakesh (156-qubit QPU) confirms reliable circuit execution with only 1.2% average deviation from simulation.
arXiv:2601.16858v2 Announce Type: replace
Abstract: The rise of generative AI as a primary information source presents a paradigm shift from traditional web search. This paper presents a large-scale empirical study quantifying the fundamental differences between the results returned by Google Search and leading generative AI services. We analyze multiple dimensions, demonstrating that AI-generated answers and web search results diverge significantly in their consulted source domains, the typology of these domains (e.g., earned media vs. owned, social), query intent, and the freshness of the information provided. We then investigate the role of LLM pre-training as a key factor shaping these differences, analyzing how this intrinsic knowledge base interacts with and influences real-time web search when enabled. Our findings reveal the distinct mechanics of these two information ecosystems, leading to critical observations on the emergent field of Answer Engine Optimization (AEO) and its contrast with traditional Search Engine Optimization (SEO).
arXiv:2605.17374v1 Announce Type: new
Abstract: The notion of software languages subsumes programming languages, modeling languages, and yet many other types of languages used in software engineering. The emerging ontology `Foundations of Software Languages' (FSL) organizes the foundations underlying software languages. We are concerned with language categories, language concepts, associated tools and methodological approaches, the formal systems or other formal entities underlying software languages, and the embedding of software languages into into software engineering activities. The primary objective of FSL is to serve as a knowledge resource in Computer Science education by connecting several subject areas in a principled manner. The first release of FSL (V1), as discussed in this paper, was built through a relatively standard methodology involving common steps for expectations, reuse, conceptualization, formalization, and validation. We leveraged GenAI to support ontology engineering (discovery, classification, linkage, completion, and transformation).
arXiv:2605.18338v1 Announce Type: cross
Abstract: Champion recommendation in multiplayer online battle arena games is usually framed informally as a problem of metagame strength, personal comfort, or global win rate. We formalize champion recommendation in League of Legends as an interpretable, player-conditional ranking problem under sparse, noisy, and non-stationary behavioral data. The proposed framework combines four information sources: a population-strength proxy, player-style similarity, direct and indirect mastery priors, and archetype-level guardrails. The method uses robust median/MAD normalization, logarithmic transforms for skewed event counts, recency-weighted player style vectors, mastery-weighted champion-pool vectors, weighted cosine similarity, rank-scaled score components, and k-means++ clustering for coarse archetype support. The implemented prototype uses a Python/Pandas modeling layer, Supabase-backed storage, and a web-facing recommendation interface. Unlike black-box supervised win-prediction systems, the proposed method returns decomposed recommendation scores that can be inspected as expected-performance proxy, fit, mastery, and archetype compatibility. A single-player case study on a 100-game history for the player identifier DIVINERAINRACCON is included as an end-to-end sanity check. The manuscript is therefore a methods and systems contribution: it specifies a reproducible, modular, and auditable champion recommender and gives a validation protocol for future large-scale evaluation through temporal train-test splits, next-champion recovery, calibration analysis, and ablation studies.
arXiv:2605.18407v1 Announce Type: cross
Abstract: While modern Large Language Models (LLMs) and agentic artificial intelligence (AI) have demonstrated transformative capabilities in digital domains, the realization of embodied AI capable of real-world scientific discovery remains a difficult frontier. The advancements are hindered by the inherent complexity of integrating high-level reasoning, multimodal information processing and real-time physical execution. Here we introduce Qumus, the first AI quantum materials experimentalist. Physically embodied within a robotic mini-laboratory, Qumus is an intelligent, multimodal, and multi-agent system designed for the creation and nano-processing of atomically thin two-dimensional (2D) materials and stacked van der Waals (vdW) structures. Qumus autonomously navigates the full scientific cycle, from hypothesis generation and protocol planning to multi-step experimental execution, result analysis and reporting, acting as an experimentalist. Markedly, the system has achieved, for the first time, the AI-creation of graphene, as well as the first AI-fabrication of complex nanodevices including atomically thin field-effect transistors via vdW stacking. Qumus excels at these tasks by demonstrating autonomous error correction and closed-loop experimentation. Our results establish a generalizable framework for self-improving embodied AI systems that learn directly from the quantum world, opening a pathway toward accelerated discovery in quantum materials, electronics and beyond.
arXiv:2605.18422v1 Announce Type: cross
Abstract: The functional ANOVA, or Hoeffding decomposition, provides a principled framework for interpretability by decomposing a model prediction into main effects and higher-order interactions. For independent inputs, this classical decomposition is explicit. It is closely connected to SHAP values, generalized additive models, and orthogonal polynomial expansions, and therefore constitutes a fundamental tool for additive explainability. In the more general and realistic dependent setting, however, obtaining a tractable representation and estimating the decomposition from data remain challenging. In this work, we address this problem for continuous inputs. By combining Hilbert space methods with the generalized functional ANOVA, we build an explicit decomposition Riesz Basis allowing to easily compute the decomposition. Our formulation recovers the classical independent case and its associated orthogonal decomposition. Building on this representation, we propose a simple but mighty algorithm to estimate the decomposition from a data sample in a model-agnostic setting and we compare it empirically with several state-of-the-art explanation methods, demonstrating the power of the approach.
arXiv:2511.04472v3 Announce Type: replace
Abstract: Malware analysis systems, including dynamic-analysis sandboxes and digital forensics and incident response (DFIR) platforms, rely on telemetry pipelines comprising collection agents, serializers, and database backends to capture and present program behavior to analysts. We show that these data-handling components constitute an exploitable attack surface that can lead to denial-of-analysis (DoA) states without disabling sensors or requiring elevated privileges. We present Telemetry Complexity Attacks (TCAs), a new class of vulnerabilities that exploit mismatches between unbounded collection mechanisms and bounded processing capabilities. Our method recursively spawns child processes to generate deeply nested and oversized objects that stress serialization and storage boundaries, as well as visualization layers, e.g., JSON/BSON depth and size limits. Depending on the product, this leads to truncated or missing behavioral reports, rejected database inserts, serializer recursion and size errors, and unresponsive dashboards, with some cases also exhibiting normal malicious execution that was not recorded or presented to analysts. We evaluate our technique against 18 commercial and open-source malware analysis platforms and endpoint detection and response (EDR) solutions. Seven products fail at different stages of the telemetry pipeline; two CVE identifiers have been assigned (CVE-61301 and CVE-61303); one more is pending; one has been assigned to an underlying library, and others have issued patches or configuration changes. We discuss root causes and propose mitigation strategies to prevent DoA attacks triggered by adversarial telemetry.
arXiv:2605.18473v1 Announce Type: cross
Abstract: Electric double-layer capacitors (EDLCs), consisting of an ionic fluid between two metallic electrodes, are electrochemical energy storage devices complementary to batteries, allowing for a faster charge/discharge. The charging dynamics in response to a voltage step features a variety of regimes and relaxation timescales, depending on the applied voltage and the various lengths characterizing the system, most importantly the inter-electrode distance and the Debye length over which electrostatic effects are screened in the electrolyte. Inspired by recent works on "shortcut to adiabaticity" in colloidal systems, here we investigate the possibility to control the charge and discharge of planar EDLCs using time-dependent voltages. Specifically, our aim is to achieve a full charge or discharge within a finite time shorter than their intrinsic relaxation timescales. Within the Poisson-Nernst-Planck model and the small-voltage regime, we derive time-dependent protocols that can eliminate an arbitrary number of relaxation modes. This permits to approach the equilibrium charged state within a finite time, that can be in practice an order of magnitude faster than the natural equilibration time. We illustrate the relevance and efficacy of the method on polynomial drivings and show that the surface charge density, charge-density profiles, and global deviation from equilibrium (quantified by a Kullback-Leibler-like divergence) can all be significantly accelerated, even for driving times comparable to or shorter than the natural RC time of the system.
arXiv:2601.05527v2 Announce Type: replace
Abstract: Accurate and efficient multivariate time series (MTS) analysis is increasingly critical for a wide range of intelligent applications. Within this realm, Transformers have emerged as the predominant architecture due to their strong ability to capture pairwise dependencies. However, Transformer-based models suffer from quadratic computational complexity and high memory overhead, limiting their scalability and practical deployment in long-term and large-scale MTS modeling. Recently, Mamba has emerged as a promising linear-time alternative with high expressiveness. Nevertheless, directly applying vanilla Mamba to MTS remains suboptimal due to three key limitations: (i) the lack of explicit cross-variate modeling, (ii) difficulty in disentangling the entangled intra-series temporal dynamics and inter-series interactions, and (iii) insufficient modeling of latent time-lag interaction effects. These issues constrain its effectiveness across diverse MTS tasks. To address these challenges, we propose DeMa, a dual-path delay-aware Mamba backbone. DeMa preserves Mamba's linear-complexity advantage while substantially improving its suitability for MTS settings. Specifically, DeMa introduces three key innovations: (i) it decomposes the MTS into intra-series temporal dynamics and inter-series interactions; (ii) it develops a temporal path with a Mamba-SSD module to capture long-range dynamics within each individual series, enabling series-independent, parallel computation; and (iii) it designs a variate path with a Mamba-DALA module that integrates delay-aware linear attention to model cross-variate dependencies. Extensive experiments on five representative tasks, long- and short-term forecasting, data imputation, anomaly detection, and series classification, demonstrate that DeMa achieves state-of-the-art performance while delivering remarkable computational efficiency.
arXiv:2605.18569v1 Announce Type: cross
Abstract: The computation of electronic excited states and real-time quantum dynamics of many-fermion systems is among the most promising applications of near-term quantum computing. In this work, we generalize the reinforcement learning contracted quantum eigensolver (RL-CQE), previously developed for ground-state problems, to electronic excited states and real-time quantum dynamics, in which a deep Q-network agent adaptively selects the two-body operators at each iteration, yielding more compact ans\"{a}tze and improved robustness with respect to critical hyperparameters. A key feature of the algorithm is a scalable state representation based on the ACSE residuals, whose dimension grows with the one-particle basis but remains independent of the number of targeted excited states. We also verify the equivalence of sign-free qubit operators in the excited-state setting, extending a result previously established for ground-state problems. Our RL-CQE for time evolution derives from a constant-scaling ansatz that represents the wave function with a fixed number of unitary transformations independent of simulation time $t$, enabled by the shared unitary structure of the purified ensemble treatment of excited states. Benchmarks on chemical systems demonstrate chemical accuracy with minimal operator counts across a range of bond lengths.
arXiv:2605.18575v1 Announce Type: cross
Abstract: Triangular-lattice systems attract a lot of attention due to various frustration-induced and strongly correlated effects. Here, we focus on the charge-ordering phenomenon by means of investigation of the extended Hubbard model with dynamical mean-field theory (DMFT). By considering the intersite nearest-neighbor interaction we have found a very rich phase diagram that contains large number of features, phases, and phase transitions. Among them are pinball-liquid (PL) phases where we distinguish charge-transfer-driven and Mott-localization-driven PLs; phase transitions that change their order as model parameters evolve (from discontinuous to continuous); very strong particle-hole asymmetry. Various features of the phase diagram are found to be better understood by means of the simple mean-field approximation (MFA). Moreover, besides helping with interpretation of the phase diagram, the MFA results together with results for the atomic-limit model are found to be able to set rather good expectations on how the DMFT phase diagram should look like. Nevertheless, a few features were not expected and are found within the DMFT, such as a small-region intermediate metallic phase on an electron-doped side of the phase diagram.
arXiv:2605.17271v1 Announce Type: cross
Abstract: Bi-causal optimal transport (OT) is a natural framework for comparing and coupling stochastic processes under nonanticipative information constraints, with important applications in robust finance, sequential uncertainty quantification, and multistage stochastic optimization. In particular, a learned bi-causal coupling naturally serves as a simulator for generating joint sample paths that respect both prescribed marginal laws and the underlying information flow. Its practical use, however, is limited by the computational difficulty of enforcing bi-causal coupling constraints over path space, especially for continuous distributions and long horizons. We develop a scalable stochastic-optimization framework for computing bi-causal OT couplings under general marginals. Our approach introduces a Kullback--Leibler (KL)-penalized relaxation that replaces hard marginal constraints with tractable divergence penalties while preserving the recursive structure of the problem. We establish dynamic programming principles for both the original and relaxed formulations, prove that the relaxed problem converges to the original bi-causal OT problem as the penalty grows, and derive explicit policy-gradient representations for the relaxed objective. Building on these results, we propose a practical policy-gradient algorithm with unbiased mini-batch estimators, variance reduction, and nonasymptotic regret guarantees. Numerical experiments show that the method accurately captures marginal laws and temporal dependence, and performs well in applications including robust subhedging and time series statistical downscaling. These results provide a scalable computational approach to bi-causal OT and broaden its applicability in settings where nonanticipative information constraints are essential.
arXiv:2604.18652v2 Announce Type: replace
Abstract: The transition of agentic AI from brittle prototypes to production systems is stalled by a pervasive crisis of craft. We suggest that the prevailing orchestration paradigm-delegating the system control loop to large language models and merely patching with heuristic guardrails-is the root cause of this fragility. Instead, we propose Arbiter-K, a Governance-First execution architecture that reconceptualizes the underlying model as a Probabilistic Processing Unit encapsulated by a deterministic, neuro-symbolic kernel. Arbiter-K implements a Semantic Instruction Set Architecture (ISA) to reify probabilistic messages into discrete instructions. This allows the kernel to maintain a Security Context Registry and construct an Instruction Dependency Graph at runtime, enabling active taint propagation based on the data-flow pedigree of each reasoning node. By leveraging this mechanism, Arbiter-K precisely interdicts unsafe trajectories at deterministic sinks (e.g., high-risk tool calls or unauthorized network egress) and enables autonomous execution correction and architectural rollback when security policies are triggered. Evaluations on OpenClaw and NanoBot demonstrate that Arbiter-K enforces security as a microarchitectural property, achieving 76% to 95% unsafe interception for a 92.79% absolute gain over native policies. The code is publicly available at https://github.com/cure-lab/ArbiterOS.
arXiv:2511.23253v3 Announce Type: replace
Abstract: Recent advancements in Vision-Language Models (VLMs) have significantly impacted various industries. In agriculture, these multimodal capabilities hold great promise for applications such as precision farming, crop monitoring, pest detection, and environmental sustainability. However, while several Visual Question Answering (VQA) datasets and benchmarks have been developed to assess VLM performance, they often fail to effectively evaluate the critical reasoning and problem-solving skills needed in complex agricultural contexts. To address this gap, we introduce AgroCoT, a VQA dataset that integrates Chain-of-Thought (CoT) reasoning, specifically designed to evaluate the reasoning capabilities of VLMs. With 4,759 carefully curated samples, AgroCoT provides a comprehensive and robust evaluation of reasoning abilities, particularly in zero-shot scenarios, focusing on the models' ability to engage in logical reasoning and effective problem-solving. Our evaluation of 30 representative VLMs, including both proprietary and open-source models, reveals a gap in their reasoning capabilities, which underscores the importance of incorporating CoT for assessments. Our dataset is available at https://huggingface.co/datasets/AgroCoT/AgroCoT.
arXiv:2511.07329v4 Announce Type: replace
Abstract: This paper proposes FractalNet, a framework based on fractal design principles that automatically generates and evaluates convolutional neural network (CNN) architectures using recursive template patterns. Rather than relying on computationally expensive Neural Architecture Search (NAS) methods, the framework explores a structured architecture space defined by recursive fractal templates that systematically vary key parameters such as fractal depth, column width, and layer configurations. The framework consists of three core components: a generator that produces candidate architectures via controlled permutations of convolutional, normalization, activation, and dropout layers; a fractal template module that enforces recursive multi-path structural patterns; and a runner module that manages model training, evaluation, and logging. Using this system, over 1,200 distinct CNN architectures were automatically generated and evaluated on the CIFAR-10 image classification benchmark. Training was performed in PyTorch using stochastic gradient descent with Automatic Mixed Precision (AMP) and gradient checkpointing to reduce computational overhead. Experimental results demonstrate that fractal-based architectures exhibit stable training dynamics and achieve competitive performance, with an average validation accuracy of 60-70% and a peak accuracy of 80.18% after only five training epochs. These findings suggest that recursive fractal structures provide an effective means of balancing network depth and width while supporting large-scale automated architecture exploration. The proposed framework offers a resource-efficient and interpretable approach to systematic neural architecture experimentation.
arXiv:2510.04309v3 Announce Type: replace
Abstract: Controlling the behaviors of large language models (LLM) is fundamental to their safety alignment and reliable deployment. However, existing steering methods are primarily driven by empirical insights and lack theoretical performance guarantees. In this work, we develop a control-theoretic foundation for activation steering by showing that popular steering methods correspond to the proportional (P) controllers, with the steering vector serving as the feedback signal. Building on this finding, we propose Proportional-Integral-Derivative (PID) Steering, a principled framework that leverages the full PID controller for activation steering in LLMs. The proportional (P) term aligns activations with target semantic directions, the integral (I) term accumulates errors to enforce persistent corrections across layers, and the derivative (D) term mitigates overshoot by counteracting rapid activation changes. This closed-loop design yields interpretable error dynamics and connects activation steering to classical stability guarantees in control theory. Moreover, PID Steering is lightweight, modular, and readily integrates with state-of-the-art steering methods. Extensive experiments across multiple LLM families and benchmarks demonstrate that PID Steering consistently outperforms existing approaches, achieving more robust and reliable behavioral control. The code is publicly available at: https://github.com/dungnvnus/pid-steering
arXiv:2508.06974v2 Announce Type: replace
Abstract: 1-bit LLM quantization offers significant advantages in reducing storage and computational costs. However, existing methods typically train 1-bit LLMs from scratch, failing to fully leverage pre-trained models. This results in high training costs and notable accuracy degradation. We identify that the large gap between full precision and 1-bit representations makes naive adaptation difficult. In this paper, we introduce a consistent progressive training for both forward and backward, smoothly converting the full-precision weights into the binarized ones. Additionally, we incorporate binary-aware initialization and dual-scaling compensation to reduce the difficulty of progressive training and improve the performance. Experimental results on LLMs of various sizes demonstrate that our method outperforms existing approaches. Our results show that high-performance 1-bit LLMs can be achieved using pre-trained models, eliminating the need for expensive training from scratch.
arXiv:2604.07649v4 Announce Type: replace
Abstract: Aggregating experimental data from papers enables materials scientists to build better property prediction models and to facilitate scientific discovery. Recently, interest has grown in extracting not only single material properties but also entire experimental measurements. To support this shift, we introduce LitXBench, a framework for benchmarking methods that extract experiments from literature. We also present LitXAlloy, a dense benchmark comprising 1426 total measurements from 19 alloy papers. By storing the benchmark's entries as Python objects, rather than text-based formats such as CSV or JSON, we improve auditability and enable programmatic data validation. We find that frontier language models, such as Gemini 3.1 Pro Preview, outperform existing multi-turn extraction pipelines by up to 0.37 F1. Our results suggest that this performance gap arises because extraction pipelines associate measurements with compositions rather than the processing steps that define a material.