arXiv:2603.22161v2 Announce Type: replace
Abstract: Metacognition -- assessing the quality of one's own cognitive performance -- guides adaptive behavior across species. Substantial research demonstrates that confidence signals can be extracted from language model outputs, yet a fundamental question remains: do models actually use these signals to control behavior, such as deciding whether to answer or abstain? To investigate, we developed a four-phase paradigm. Phase~1 elicited baseline confidence estimates without an abstention option. Phase~2 revealed that LLMs apply an implicit threshold to internal confidence when deciding to abstain, with confidence effect sizes approximately an order of magnitude larger than alternative mechanisms. Phase~3 provided direct causal evidence through activation steering: boosting or suppressing confidence signals correspondingly decreased or increased abstention rates. Phase~4 extended this by systematically varying instructed thresholds, demonstrating that LLMs actively deploy confidence signals to implement abstention policies. Critically, beyond calibrated log-probability based confidence derived from the output distribution, verbal confidence independently predicted abstention across all models, despite being objectively less discriminatory of answer correctness. Activation decoding at the last pre-answer token further showed that both observable measures are lossy readouts of a richer internal representation. Together, these results suggest that abstention is not fully captured by the strength of evidence in the output distribution alone, but is better explained by the joint operation of a multidimensional internal confidence representation and threshold-based policies -- consistent with structured metacognitive control in LLMs, a capacity of growing importance as models transition to autonomous agents that must recognize their own uncertainty.
Science Journals
arXiv:2603.22453v2 Announce Type: replace
Abstract: Community Notes have emerged as an effective crowd-sourced mechanism for combating online deception on social media platforms. However, its reliance on human contributors limits both the timeliness and scalability. In this work, we study the automated Community Notes generation task for image-based contextual deception, where an authentic image is paired with misleading context (e.g., time, entity, and event). Unlike prior work that primarily focuses on deception detection (i.e., judging whether a post is true or false in a binary manner), automated Community Notes generation requires producing concise and grounded notes that help users recover the missing or corrected context. This problem remains underexplored due to the scarcity of datasets that support this task. To address this gap, we curate a real-world dataset, XNote, comprising X posts with associated Community Notes and external contexts, along with annotations of topics and deceptive factors. We further benchmark a range of frontier large vision language models (LVLMs) on XNote, evaluating their performance on both deception detection and note generation tasks. We also compare against an end-to-end approach, SNIFFER, and a commercial tool, GPT-5. Our results highlight the challenges in automated Community Notes generation, underscoring the need for improved methods and metrics tailored for this task.
arXiv:2605.19386v1 Announce Type: new
Abstract: Reconstructing simulation-ready deformable objects is important for vision, graphics, and robotics. Existing physics-driven methods can recover physical digital twins from videos, but they suffer from two fundamental limitations: they typically assume a homogeneous material across the whole object, and their scene-specific inverse optimization, combined with the inherent ambiguity of monocular observation, yields inconsistent parameters for the same material across different scenes or interactions. We propose MatPhys, a material-aware feed-forward framework that predicts spring-mass parameters from a single-view video, addressing these two issues with two coupled designs. To relax the homogeneous material assumption, we use DINO features to decompose the object into semantically meaningful parts and to query a part-level material prior, assigning each part its own physical behavior. To enforce cross-scene consistency, we introduce a learned material codebook of shared material embeddings as the bridge between appearance and physics, and further use the part-level prior as a reference distribution that constrains the decoder so that the same material yields consistent parameters across scenes and interactions. Together, these designs turn an under-constrained monocular problem into feed-forward inference grounded on shared, reusable material concepts. Experiments show that our method matches per-scene optimization baselines in reconstruction and future prediction, while achieving stronger generalization to unseen interactions and objects with more consistent physical parameters.
arXiv:2505.11628v4 Announce Type: replace
Abstract: Supervised fine-tuning with expert demonstrations often produces models that imitate outputs without internalizing the reasoning processes needed for robust generalization. While critique-based approaches show promise, training models to generate critiques directly, such as Critique Fine-Tuning (CFT), can lead to output-format drift and degradation of general capabilities. We propose Critique-Guided Distillation (CGD), a training framework that decouples critique consumption from critique generation. During fine-tuning, the student is trained to refine flawed responses conditioned on teacher critiques. CGD treats critiques as a \textit{training-time-only} supervision signal, encouraging internalization of error-aware reasoning: critiques guide learning but are absent at inference. Controlled ablations confirm that these reasoning gains are directly driven by the specificity and relevance of the teacher's feedback. Across five model families, CGD consistently outperforms CFT and standard distillation on mathematical reasoning benchmarks, yielding 7\% average improvements and gains of up to +15.0\% on AMC23 and +12.2\% on MATH-500. On challenging competition problems such as AIME24 and AIME25, CGD achieves substantially higher Pass@1 and stronger performance at low Pass@k, indicating improved reasoning quality per sample. Importantly, CGD preserves general instruction-following capabilities where CFT degrades significantly ($-$21.3\% on IFEval). These results position CGD as a practical and compute-efficient intermediate training paradigm for reasoning-centric tasks without introducing architectural inference-time overhead.
arXiv:2605.19779v1 Announce Type: new
Abstract: We adapt split conformal prediction and adaptive conformal inference (ACI) to continuous AI agent evaluation, providing distribution-free coverage guarantees for forecasted quality scores. Conformal intervals achieve calibration error below 0.02 across all nominal levels at the 24h horizon, while ACI correctly widens intervals by 35% following agent releases then reconverges. We further develop compositional uncertainty bounds for multi-agent pipelines (validated via simulation across inter-stage correlations rho in [-0.5, 0.9]), a conformal abstention rule for pairwise rankings with controlled false-ranking rate, and FDR-corrected abstention for leaderboard-scale multiple testing. Evaluating 50 agents via 18 real-time signals collected hourly, we show that per-agent conditional coverage is well-concentrated around the nominal level (mean 80.4%, 90% of agents within [72%, 90%]), and that cross-source sentiment divergence predicts ranking instability (r=0.64, p<0.01). A circularity-controlled validation confirms the framework captures signal beyond benchmarks (rho_s=0.52, p<0.01, n=35). Code and data are released under CC BY 4.0.
arXiv:2605.19407v1 Announce Type: new
Abstract: We investigate data filtering for large model pretraining via new scaling studies that target the high compute, data-scarce regime. In spite of an apparently common belief that filtering data to include only high-quality information is essential, our experiments suggest that with enough compute, the best data filter is no data filter. We find that sufficiently trained large parameter models not only tolerate low-quality and distractor data, but in fact benefit from nominally ``poor'' data.
arXiv:2505.23747v2 Announce Type: replace
Abstract: Recent advancements in Multimodal Large Language Models (MLLMs) have significantly enhanced performance on 2D visual tasks. However, improving their spatial intelligence remains a challenge. Existing 3D MLLMs always rely on additional 3D or 2.5D data to incorporate spatial awareness, restricting their utility in scenarios with only 2D inputs, such as images or videos. In this paper, we present Spatial-MLLM, a novel framework for visual-based spatial reasoning from purely 2D observations. Unlike conventional video MLLMs which rely on CLIP-based visual encoders optimized for semantic understanding, our key insight is to unleash the strong structure prior from the feed-forward visual geometry foundation model. Specifically, we propose a dual-encoder architecture: a pretrained 2D visual encoder to extract semantic features, and a 3D spatial encoder-initialized from the backbone of the visual geometry model-to extract 3D structure features. A connector then integrates both features into unified visual tokens for enhanced spatial understanding. Furthermore, we propose a space-aware frame sampling strategy at inference time, which selects the spatially informative frames of a video sequence, ensuring that even under limited token length, the model focuses on frames critical for spatial reasoning. Beyond architecture improvements, we construct a training dataset from multiple sources and train the model on it using supervised fine-tuning and GRPO. Extensive experiments on various real-world datasets demonstrate that Spatial-MLLM achieves state-of-the-art performance in a wide range of visual-based spatial understanding and reasoning tasks. Project page: https://diankun-wu.github.io/Spatial-MLLM/.
arXiv:2506.00286v3 Announce Type: replace
Abstract: We study risk-sensitive reinforcement learning in finite discounted MDPs with recursive entropic risk measures (ERM), where the risk parameter $\beta \neq 0$ controls the agent's risk attitude: $\beta>0$ for risk-averse and $\beta<0$ for risk-seeking behavior. A generative model of the MDP is assumed to be available. Our focus is on the sample complexities of learning the optimal state-action value function (value learning) and an optimal policy (policy learning) under recursive ERM. We introduce a model-based algorithm, called Model-Based ERM $Q$-Value Iteration (MB-RS-QVI), and derive PAC-type bounds on its sample complexity for both value and policy learning. Both PAC bounds scale exponentially with $|\beta|/(1-\gamma)$, where $\gamma$ is the discount factor. We also establish corresponding lower bounds for both value and policy learning, showing that exponential dependence on $|\beta|/(1-\gamma)$ is unavoidable in the worst case. The bounds are tight in the number of states and actions ($S$ and $A$), providing the first rigorous sample complexity guarantees for recursive ERM across both risk-averse and risk-seeking regimes.
arXiv:2605.19408v1 Announce Type: new
Abstract: Manipulating nanoscale light-matter interactions on ultrafast time scales is indispensable for future polaritonic devices. Hyperbolic phonon polaritons (HPhPs) in van der Waals materials enable deep subwavelength confinement of electromagnetic fields in the infrared region and long-distance propagation of polaritonic waves. However, achieving ultrafast imaging and optical control of HPhPs remains a major challenge. Here, we demonstrate the direct observation of transient modulation of HPhPs induced by local photocarrier generation in WS$_2$/hBN heterostructures using ultrafast infrared scanning near-field optical microscopy. We implement grating-based spectral filtering of broadband near-field scattering to simultaneously achieve nanoscale and femtosecond spatiotemporal resolution together with fine spectral selectivity. This ultrafast nano-imaging technique reveals that photocarriers in WS$_2$ modulate the polaritonic field amplitudes and wavelengths of HPhPs in hBN. Theoretical simulations corroborate that these changes arise from photoinduced changes in WS$_2$ dielectric properties. This approach offers a versatile platform for exploring ultrafast polaritonic dynamics at the nanoscale.
arXiv:2605.19412v1 Announce Type: new
Abstract: Program reduction is a technique for simplifying large, failure-inducing programs into minimal reproducible test cases. Language-specific tools such as CReduce achieve strong performance by leveraging deep semantic knowledge of C/C++, but are tightly coupled to a single language family. Language-agnostic reducers such as Perses address this by applying syntax-guided search across any grammar, yet share a fundamental limitation: deleting a node or subtree in isolation often breaks semantic coherence causing the property checker to reject the deletion and forcing the reducer to backtrack, limiting overall reduction effectiveness and efficiency. In this paper, we propose DRReduce, a framework that bridges this gap by augmenting language-agnostic syntactic reduction with a lightweight semantic layer: dependency reconstruction, which repairs program dependencies broken by a deletion in order to preserve the semantic validity of intermediate programs and increase the acceptance rate of the property checker. DRReduce constructs a semantic dependency graph from the input program, performs semantically coherent deletions with dependency reconstruction, and delegates further minimization to a syntax-guided reducer. We implement DRReduce for C and Java and evaluate it on real-world bug-triggering programs. Compared to SOTA syntax-guided reducers, DRReduce achieves average size reductions of 51.9%, 14.9%, and 19.8% over Perses, WDD, and CDD respectively, while completing reduction faster on the majority of programs. Compared to language-specific tools, DRReduce achieves results comparable to CReduce and Latra without any language-specific transformation rules, at 3.3x and 1.2x higher efficiency than CReduce and Latra on average, respectively. An ablation study confirms that dependency reconstruction reduces query invocations by 80.2%, reduction time by 58.7%, and final token count by over 55.1%.
arXiv:2603.25620v4 Announce Type: replace
Abstract: Large language model (LLM)-based persona agents are rapidly being adopted as scalable proxies for human participants across diverse domains. Yet there is no systematic method for verifying whether a persona agent's responses remain free of contradictions and factual inaccuracies throughout an interaction. A principle from interrogation methodology offers a lens: no matter how elaborate a fabricated identity, systematic interrogation will expose its contradictions. We apply this principle to propose PICon, an evaluation framework that probes persona agents through logically chained multi-turn questioning. PICon evaluates consistency along three core dimensions: internal consistency (freedom from self-contradiction), external consistency (alignment with real-world facts), and retest consistency (stability under repetition). Evaluating seven groups of persona agents alongside 63 real human participants, we find that even systems previously reported as highly consistent fail to meet the human baseline across all three dimensions, revealing contradictions and evasive responses under chained questioning. This work provides both a conceptual foundation and a practical methodology for evaluating persona agents before trusting them as substitutes for human participants. We provide the source code and an interactive demo at: https://kaist-edlab.github.io/picon/
arXiv:2605.19425v1 Announce Type: new
Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has become the dominant paradigm for advanced reasoning in Large Language Models (LLMs), but rollout samples are expensive to obtain, making sample efficiency a critical bottleneck. A natural remedy is to reuse each rollout batch for multiple gradient updates, a standard practice in classical RL. Yet in RLVR, this amplifies policy shift, leading to severe performance degradation. Detecting the onset of degradation early enough to stop reuse remains an open and challenging problem. We close this gap by identifying the \textit{Disproportionate Weight Divergence (DWD)} phenomenon: performance degradation is synchronized with a sharp surge in the \texttt{lm\_head} weight change, while intermediate layers remain stable. Empirically, we verify that DWD emerges consistently across diverse LLMs and tasks. Theoretically, we prove that (i) harmful gradients concentrate at the \texttt{lm\_head} while intermediate layers are structurally attenuated, and (ii) the \texttt{lm\_head} gradient norm lower-bounds the policy divergence. These results establish the \texttt{lm\_head} gradient norm as a principled, real-time signal of catastrophic policy shift. Guided by this insight, we propose \textit{Dynamic Gradient Gating (DGG)}, a lightweight intervention that monitors the \texttt{lm\_head} gradient norm in real time and intercepts harmful gradients before they corrupt the optimizer. DGG consistently matches or exceeds the standard single-use baseline, achieving up to $2.93\times$ sample efficiency and $2.14\times$ wall-clock speedup across math, ALFWorld, WebShop, and search-augmented QA tasks.
arXiv:2506.14148v2 Announce Type: replace
Abstract: This paper presents a novel non-invasive object classification approach using acoustic scattering, demonstrated through a case study on hair assessment. When an incident wave interacts with an object, it generates a scattered acoustic field encoding structural and material properties. By emitting acoustic stimuli and capturing the scattered signals from head-with-hair-sample objects, we classify hair type and moisture using AI-driven, deep-learning-based sound classification. We benchmark comprehensive methods, including (i) fully supervised deep learning, (ii) embedding-based classification, (iii) supervised foundation model fine-tuning, and (iv) self-supervised model fine-tuning. Our best strategy achieves nearly 90% classification accuracy by fine-tuning all parameters of a self-supervised model. These results highlight acoustic scattering as a privacy-preserving, non-contact alternative to visual classification, opening huge potential for applications in various industries.
arXiv:2506.18852v2 Announce Type: replace
Abstract: Mechanistic interpretability (MI) aims to explain how neural networks work by uncovering their underlying mechanisms. As the field grows in influence, it is increasingly important to examine not just models themselves, but the assumptions, concepts and explanatory strategies implicit in MI research. We argue that mechanistic interpretability needs philosophy as an ongoing partner in clarifying its concepts, refining its methods, and navigating the epistemic and ethical complexities of interpreting AI systems. There is significant unrealised potential for progress in MI to be gained through deeper engagement with philosophers and philosophical frameworks. Taking three open problems from the MI literature as examples, this paper illustrates the value philosophy can add to MI research, and outlines a path toward deeper interdisciplinary dialogue.
arXiv:2605.19446v1 Announce Type: new
Abstract: Recently, pre-trained encoders have gained widespread use due to their strong capability in representation extraction. However, they are vulnerable to downstream-agnostic attacks (DAAs). Existing DAA methods operate under a permissive threat model, where an attack is successful if the generated downstream-agnostic adversarial examples (DAEs) change the original prediction, without requiring a specific target. In this paper, we propose a Targeted DAA (TDAA) method under a stricter threat model requiring the attack to be both targeted and downstream-agnostic. Since the downstream task is unknown and encoders do not directly produce predictions, achieving a targeted attack is particularly challenging. To address this, we introduce a novel component termed the 'threat image', pre-selected by the attacker as the target. Specifically, a generator is designed to produce example-specific adversarial perturbations that compel the victim encoder to output identical features for both the DAEs and the threat image. Unlike previous DAA methods that generate a single shared perturbation for all samples, which often fails due to image diversity, our method adopts an example-specific paradigm. This generates tailored perturbations for each image to ensure a high attack success rate and invisibility. By leveraging the threat image as a feature-level anchor, our method builds a task-agnostic bridge to reveal the vulnerabilities of the victim encoder. Extensive experiments on 10 self-supervised methods across 3 benchmark datasets demonstrate the effectiveness of our approach and reveal the pronounced vulnerability of pre-trained encoders. The code will be made publicly available after the review period.
arXiv:2604.07035v2 Announce Type: replace
Abstract: Open reasoning language models are often compared under mixed sample sizes, partially standardized prompts, and accuracy-centered summaries, which makes practical model selection difficult to interpret. We present a unified evaluation of seven open reasoning language model configurations across four benchmarks: ARC-Challenge, GSM8K, MATH levels 1 to 3, and TruthfulQA MC1. We test zero-shot, chain-of-thought (CoT), and few-shot CoT prompting on the same 238-example subset for every model--dataset--strategy condition, yielding a complete 7 x 4 x 3 design with 84 conditions and 19,992 evaluated examples. Beyond accuracy, we report Wilson confidence intervals, latency, peak video random access memory (VRAM), weighted aggregate performance, Pareto-efficient operating points, prompt-sensitivity metrics, and compatibility diagnostics. Gemma-4-26B-A4B with zero-shot prompting achieves the highest weighted score at 0.794. Gemma-4-E4B remains close to the top across prompting settings while using substantially lower latency and memory, making it a strong practical operating point. Bootstrap and paired-permutation analyses show that the leading configurations are close enough that deployment tradeoffs remain important. We also find that prompting strategy changes model rankings rather than shifting all models uniformly. Benchmark-specific complementarity creates routing headroom, with an oracle task-aware selector reaching a weighted score of 0.825. Compatibility diagnostics show that some apparent failures, especially Phi-4-Reasoning on GSM8K, reflect robustness and interface-adherence problems under the shared evaluation pipeline. These results support a central claim: open-model evaluation should be framed as a deployment-aware, multi-objective operating-point problem rather than as a single-score leaderboard exercise.
arXiv:2604.14348v2 Announce Type: replace
Abstract: Urban venues serve as arenas for social mixing. While residential and activity-space segregation have been extensively studied, less is known about how the spatial structure of cities, particularly public transit infrastructure, shapes the geography of social mixing at specific locations. This study examines how transit accessibility associates with visitor diversity -- the compositional heterogeneity of visitors sharing a venue, used here as an indicator of social mixing potential -- at points of interest (POIs) in nine cities in Sweden and three cities in the United States (New York, Washington DC, Atlanta). Using mobile phone GPS data in 2024, we compute visitor diversity indices based on the birth background composition of visitors' home neighborhoods. Transit catchment diversity positively predicts visitor diversity, but this association is robust only in the largest metropolitan areas; in smaller Swedish cities, the coefficient attenuates to insignificance once geographic catchment composition, centrality, and venue density are controlled. Transit-diversity hotspots concentrate not in already diverse venues, but in lower-diversity POIs with lower commercial density, greater distance from transit in US cities, and greater centrality in Sweden. These patterns are consistent with transit infrastructure playing a bridging role, linking diverse populations to venues where alternative pathways are limited.
arXiv:2409.03192v2 Announce Type: replace
Abstract: Fine-grained image classification has witnessed significant advancements with the advent of deep learning and computer vision technologies. However, the scarcity of detailed annotations remains a major challenge, especially in scenarios where obtaining high-quality labeled data is costly or time-consuming. To address this limitation, we introduce Precision-Enhanced Pseudo-Labeling(PEPL) approach specifically designed for fine-grained image classification within a semi-supervised learning framework. Our method leverages the abundance of unlabeled data by generating high-quality pseudo-labels that are progressively refined through two key phases: initial pseudo-label generation and semantic-mixed pseudo-label generation. These phases utilize Class Activation Maps (CAMs) to accurately estimate the semantic content and generate refined labels that capture the essential details necessary for fine-grained classification. By focusing on semantic-level information, our approach effectively addresses the limitations of standard data augmentation and image-mixing techniques in preserving critical fine-grained features. We achieve state-of-the-art performance on benchmark datasets, demonstrating significant improvements over existing semi-supervised strategies, with notable boosts in accuracy and robustness.
arXiv:2604.15034v4 Announce Type: replace
Abstract: Recent advances in LLM based agent systems have shown promise in tackling complex, long horizon tasks. However, existing agent protocols (e.g., A2A and MCP) under specify cross entity lifecycle and context management, version tracking, and evolution safe update interfaces, which encourages monolithic compositions and brittle glue code. We introduce Autogenesis Protocol (AGP), a self evolution protocol that decouples what evolves from how evolution occurs. Its Resource Substrate Protocol Layer (RSPL) models prompts, agents, tools, environments, and memory as protocol registered resources with explicit state, lifecycle, and versioned interfaces. Its Self Evolution Protocol Layer (SEPL) specifies a closed loop operator interface for proposing, assessing, and committing improvements with auditable lineage and rollback. Building on AGP, we present Autogenesis System (AGS), a self-evolving multi-agent system that dynamically instantiates, retrieves, and refines protocol-registered resources during execution. We evaluate AGS on multiple challenging benchmarks that require long horizon planning and tool use across heterogeneous resources. The results demonstrate consistent improvements over strong baselines, supporting the effectiveness of agent resource management and closed loop self evolution. The code is available at https://github.com/DVampire/Autogenesis.
arXiv:2604.20575v3 Announce Type: replace
Abstract: We prove that every fan-in $2$ noncommutative arithmetic circuit computing the palindrome polynomial has size $\Omega(nd)$. In particular, when $d=n$ we obtain an $\Omega(n^2)$ lower bound. The proof builds on and refines a previous work of the author. Key ideas in the proof were generated by Gemini 3.1 Pro.
arXiv:2602.09872v2 Announce Type: replace
Abstract: Human activity recognition (HAR) on resource constrained devices requires high accuracy across diverse sensor setups. Selective state space models (SSMs) offer efficient linear time sequence processing, presenting a compelling alternative to attention mechanisms. However, their TinyML design space remains unexplored. This paper introduces BabyMamba-HAR, comprising two lightweight architectures: (1) CI-BabyMamba-HAR, utilizing a channel independent stem for noise robustness, and (2) Crossover-BiDir-BabyMamba-HAR, utilizing an early fusion stem for channel count independent complexity. Both integrate weight tied bidirectional scanning and gated temporal attention pooling. Across eight benchmarks, Crossover-BiDir-BabyMamba-HAR averages an 86.52% F1-score with 27K parameters and 2.21M MACs, matching TinyHAR (86.16%) while requiring 11x fewer MACs on high channel datasets. On-device deployment on the Raspberry Pi Pico 2 and ESP32 utilized a mixed precision C++ runtime (INT8 projections, float32 states). A fused computation strategy with lifetime aware memory management reduces peak memory footprint from O(B*dmodel*L*dstate) to O(B*dmodel*dstate), adapting to support weight-tied bidirectional and channel-streaming execution. Both architectures achieved full 8/8 dataset coverage with >99.2% PyTorch parity, whereas INT8 quantized TFLite baselines showed degraded coverage and parity (TinyHAR: 7/8 and 4/8 coverage at 60.4% and 88.6% parity, TinierHAR: 8/8 and 6/8 at 54.2% and 90.8%, DeepConvLSTM: 1/8 and 0/8 on Pico 2 and ESP32, respectively). Crossover-BiDir-BabyMamba-HAR averages 154.4 ms latency on ESP32 and 481.9 ms on Pico 2. Ablations confirm bidirectional scanning and gated attention improve F1-scores by up to 8.42% and 8.94%, respectively, establishing practical principles for TinyML SSM deployment.
arXiv:2404.01063v2 Announce Type: replace
Abstract: Bioscientists frequently seek to visualize the biological systems they have empirically characterized and reported in the literature. Realizing such visualizations requires biological structure modeling, an inherently complex process that demands both biological and geometric understanding. This paper addresses the problem of constructing such 3D models for visualization. In this paper, we introduce a novel agent framework that mitigates the challenges of operating 3D modeling software by transforming user inputs, including natural language descriptions, research publication content, and textual descriptions of the existing objects and structures in the current scene, into modeling operations in a structured JSON format and final 3D results. The major technical contribution lies in the collaborative agent design that simultaneously supports model planning, execution, and novel user interaction design, such as interactive modeling execution and dynamic widget generation that fuse text and mouse interaction within the chat window. The framework further incorporates a customized modeling memory to enhance user interaction, featuring components such as personalized memory management, feedback collection, and skill library design. This modeling memory is leveraged to enable improved 3D modeling performance over time. The quantitative evaluation on our collected dataset showcases the effectiveness of our framework. We also develop a prototype tool, Chat Modeling, and demonstrate its usage through two modeling case studies. Our user study and expert interviews highlight the potential of our approach for use in scientific workflows.
arXiv:2402.01411v3 Announce Type: replace
Abstract: Context: LLM-based multi-agent systems enable automation and decision support in software development, yet existing studies rely on benchmark datasets offering only binary pass-or-fail results, limiting insight into real-world applicability. Objective: This study empirically investigates the potential and limitations of LLM-based agents in autonomous software development tasks. Method: A two-phase approach was employed: developing a multi-agent system, CodePori, for automated code generation, and conducting participant-based evaluation to assess practical performance. Results: Participant feedback reveals key strengths, challenges, and areas for improvement in LLM-based multi-agent systems, highlighting aspects missed by standard code-generation benchmarks. Conclusions: While LLM-based multi-agent systems show potential for large-scale software development, successful integration requires addressing challenges such as memory limitations, hallucinations, and code smells, alongside a practitioner-centric perspective.
arXiv:2605.20076v1 Announce Type: cross
Abstract: In this paper we propose a novel unified cosmological model that connects a pre-inflationary epoch, starting at the Planckian time, with the onset of inflation within a single scalar-field framework. The pre-inflationary phase is characterized by a decelerated expansion with an increasing comoving Hubble horizon, followed by a gradually transition to an accelerated inflationary regime. This early dynamics leads to a modified causal structure that naturally accounts for the suppression of large-angle $(\theta \gtrsim 60^\circ)$ correlations in the cosmic microwave background (CMB) reported by the satellite PLANCK. We study the quantum fluctuations of the scalar field using the Mukhanov-Sasaki formalism and a canonical quantization procedure based on energy minimization. We find that the vacuum state is well-defined only for sub-horizon modes at the onset of inflation, which induces a natural cutoff in the primordial power spectrum. The resulting spectrum exhibits a suppression at large scales while remaining nearly scale-invariant at small scales. In the appropriate limit, the model recovers the standard de Sitter result, in agreement with current observational constraints. These results highlight the relevance of pre-inflationary dynamics for addressing large-scale anomalies within a consistent inflationary framework.
arXiv:2605.19610v1 Announce Type: cross
Abstract: We investigate the asymptotic properties of the L\'evy Adaptive B-spline (LABS) regression model, a Bayesian nonparametric method that incorporates B-spline kernels into the L\'evy Adaptive Regression Kernel (LARK) model. LABS applies splines of varying degrees with independently defined knots, yielding a flexible model class capable of adapting to irregular and locally structured features of the true function. Within the nonparametric regression framework with univariate random design and Gaussian errors, we establish that the LABS posterior contracts around the true function in Besov classes at nearly minimax-optimal rates, up to a logarithmic factor, while adapting automatically to unknown smoothness. This study contributes to filling a gap in the literature, where theoretical results on posterior contraction of the LARK model in Besov spaces remain scarce. Simulation experiments on standard test functions in Besov spaces, including Blocks, Bumps, HeaviSine, and Doppler, complement the theoretical results and demonstrate the practical utility of LABS.