Artificial Intelligence
Is this machine playing?
Nathan Cloos, Antonio Norelli, Daniel Durbin et al. (+3 more)
We placed a modern AI coding assistant in an unintended role: as the mind of a body on an unknown digital island. With only a minimal instruction mentioning no specific task, reward, or activity, the machine started animating its virtual body. Across thirty-hour runs, the
embodied AI
agent climbed hills, stacked blocks into towers, drew mandalas, reinterpreted sports, ran experiments on the physics of its world, and learned techniques that later expanded what it could accomplish. These activities recurred across thirteen
agents but diverged into distinct histories. We examine whether this behavior satisfies classical criteria for play and ask whether play can become a mode of machine development.
Read the full abstract ▾
All authors Nathan Cloos, Antonio Norelli, Daniel Durbin, Jacob Andreas, Daniela Rus, Phillip Isola
Social and Information Networks
Aggregating User Preferences while Ensuring Equity, Diversity, and Inclusion using Graph Summarization
Adji Marieme Sita Cissé and Malek Mouhoub
Aggregating the preferences of diverse user groups into a collective outcome raises fundamental challenges of equity, diversity, and inclusion (EDI): classical aggregation rules such as Borda and Condorcet have no mechanism to prevent results from systematically favoring majority groups, collapsing onto homogeneous items, or under-representing minorities. We address this problem through EDI-constrained graph summarization. User preferences are modeled as a weighted attributed bipartite graph, and a greedy coarsening
algorithm iteratively merges user nodes while enforcing three structural EDI criteria: an equity gap constraint ($\Delta E$), an intra-list diversity constraint (ILD), and a group inclusion constraint. Rather than correcting
fairness after aggregation, our method embeds EDI preservation directly into the graph structure. We evaluate across five
datasets spanning four domains: MovieLens 100k and 1M, libimseti.cz, Rate My Professors, and OpenAlex (2018-2023). Our method, AURORA, achieves the largest and most consistent diversity gains over classical voting rules, and on MovieLens 100k at $k = 20$ it simultaneously improves all three EDI criteria over both Borda and Condorcet. On Rate My Professors, it combines high diversity (ILD = 0.808) with the highest female item representation (60%), at a moderate equity cost, and it achieves the lowest equity gap ($\Delta E = 0.031$) on OpenAlex, where Borda-based methods recommend zero female authors. On libimseti.cz, the only
dataset where the sensitive attribute is present on both sides of the bipartite graph, our method does not reduce the equity gap, a limitation we connect to prior findings that demographic parity is not always an appropriate target. These results demonstrate that
embedding EDI constraints into aggregation structure yields more robust
fairness-diversity trade-offs than post-hoc approaches.
Read the full abstract ▾
Computer Vision
PlaySuite: A Large-Scale Benchmark for Interactive Visual Intelligence
Dheeraj Varghese, Anna Vettoruzzo, Walter Simoncini et al. (+5 more)
Recent advances in
multimodal foundation models yield strong performance on static perception and
reasoning benchmarks, yet such evaluations largely overlook a central aspect of intelligence: acting competently in dynamic environments over extended time horizons. We introduce PlaySuite, a large-scale
benchmark for evaluating interactive visual intelligence across more than 5K open-source
video games curated from PyWeek and itch.io. Spanning diverse genres and engines, including Pygame, HTML5, Godot, and Unity, these independent games are largely out-of-distribution for current models, reducing the likelihood that success can be achieved by retrieving memorized walkthroughs or web-scale training artifacts. To enable scalable evaluation across heterogeneous titles, we develop a unified closed-loop interaction framework optimized for HPC clusters alongside a
Video-LLM-as-a-judge protocol that maps observable gameplay milestones to standardized progress levels. We evaluate fourteen recent open models spanning
vision-
language models, computer-use
agents, and
vision-
language-action models. Our results yield strong evidence of a perception-action gap: despite strong
reasoning capabilities, current models struggle to make sustained progress and exhibit recurring failures in spatial grounding, action execution, and self-correction. PlaySuite provides a reproducible and extensible testbed for measuring progress from visual perception to goal-directed interaction, and a foundation for developing models that can act, adapt, and generalize in dynamic visual environments.
Read the full abstract ▾
All authors Dheeraj Varghese, Anna Vettoruzzo, Walter Simoncini, Michelle Lorena Acevedo Callejas, Mohammad Mahdi Derakhshani, Kristof Meding, Joaquin Vanschoren, Cees G. M. Snoek
Distributed Computing · Cross-listed from math.NT
CUDA-MPQS: A GPU-Resident Self-Initializing Quadratic Sieve, and the Factorization of RSA-155
Fabian Januszewski and Christoph Heinrichs
The quadratic sieve (QS) is an irregular, cache-hostile, branch-heavy integer factorization
algorithm; prior GPU work accelerates individual stages of it. We present \swname{}, a self-initializing quadratic sieve in which every stage —
polynomial initialization, sieving, relation post-processing, large-
prime matching, \GFtwo{}
matrix construction, Block Wiedemann linear
algebra and the square root — executes on the GPU, the CPU restricted to orchestration, setup and I/O, the sieve window 99.9% GPU-busy with no device-wide or event synchronization and no synchronous copy. With it we factored the 512-bit RSA-155 challenge modulus end to end on GPUs: a 64$\times$H100 sieve feeding a single-H100 \GFtwo{} solve, 700.6~GPU-h and 242~kWh in total, 98.4% of it sieving. To our knowledge this is the largest integer factored by the quadratic sieve, exceeding the prior QS record RSA-150 (Buhrow, YAFU, 2025); all comparable general factorizations since 1996, RSA-155's own in 1999 included, used the number field sieve. On a single device we factor RSA-100 in 29.2~s on an H100 and 51.0~s on a consumer RTX~5070~Ti; in a controlled same-modulus measurement, with energy measured on both sides, the H100 is faster than the fastest CPU quadratic sieve, YAFU, on 96 cores of an EPYC~9655 (106–122~s), and than CADO-NFS's SIQS (292~s) and GNFS (263~s). Across four discrete GPUs (a 3.7$\times$ range in CUDA cores) RSA-100 throughput is close to proportional to core count,
memory bandwidth being a weaker predictor, and measured cost over seven
benchmark factorizations from RSA-100 to RSA-155 grows as $\LN{1/2}{1}^{1.04}$ ($R^2=0.993$). On RSA-150, the identical modulus on which the prior QS record was set, we measure 302.9~GPU-h against that run's 11{,}664 core-hours.
Read the full abstract ▾
Cryptography and Security
Jailbreaking Open-Weight LLMs via Random Embedding Perturbations
Abhinav Sudhakar Dubey, Scott Sirri, Vaggos Chatziafratis et al. (+1 more)
While open-weight models have enjoyed steady progress in capabilities and wide adoption across multiple domains, their safety remains an important concern. One key feature is the ability to refuse or deflect harmful, malicious, or insensitive
prompts. In this paper, we expose safety vulnerabilities across six common open-weight LLMs of various sizes that consistently lead to harmful or unsafe responses on the JailbreakBench
benchmark dataset. Our proposed attack, Perturbed
Embedding Vector (PEV), is a simple and fast "jailbreaking" technique that is cheaper than prior approaches, which typically require gradient computations, per-
prompt optimizations, or altering internal weights of the models. PEV just adds independent Gaussian noise in the
embedding vector representations of the
prompt, with no need for further manipulations. To generate unsafe responses, we repeatedly sample additive noise from this distribution. In our experiments, we observe that the average compute cost to get the first successful attack is up to an order of magnitude less than previous attacks. The first successful jailbreak on a new
prompt typically arrives within one minute on every tested model, and PEV generates unsafe responses across all models for all
prompts in JailbreakBench. No other tested method achieves such results, despite them taking longer to run. More broadly, we believe that understanding the behavior of LLMs under perturbations in the
embedding vectors is an important research direction: while perturbations constitute a major security risk, they can also serve as a valuable tool for exploring the dynamical behavior of such models.
Read the full abstract ▾
All authors Abhinav Sudhakar Dubey, Scott Sirri, Vaggos Chatziafratis, C. Seshadhri
Computational Complexity · Cross-listed from quant-ph
Black Hole Radiation Decoding in the Haar Random Oracle Model
Ezekiel Cochran and Atul Mantri
We prove optimal query bounds for recovering a single
qubit from black-hole radiation in the Haar random oracle model, when the remaining black hole contains at most one sixteenth of the system's
qubits. Recovery with any constant Haar-averaged advantage over trivial decoding requires queries proportional to the
Hilbert-space dimension of the remaining black hole. The lower bound is unconditional for decoders chosen independently of the sampled
unitary, allows arbitrary computation between queries, and access to $U, U^\dagger, U^*, U^\mathsf T$ along with the controlled variants. An existing decoder using only forward and inverse queries attains a matching bound. The proof uses a path recording oracle to compare real and maximally mixed states. As applications, we obtain an efficiently preparable, statistically far, computationally indistinguishable (EFI) pair and
quantum commitments relative to a public Haar oracle, as well as prove a tight linear rank lower bound for Uhlmann transformation on a fixed-target family.
Read the full abstract ▾
Formal Languages and Automata
Irreducibility of Endomorphisms of Finitely Generated Free Semigroups
Paul C. Bell, Eva Foster and Daniel Reidenbach
We introduce and investigate the irreducibility of endomorphisms of finitely generated free
semigroups, i.e., we investigate when an endomorphism $\phi: \Sigma^+ -> \Sigma^+$, where $\Sigma$ is any alphabet, can be nontrivially expressed as a composition $\phi = \psi_2 \circ \psi_1$ of endomorphisms $\psi_1, \psi_2: \Sigma^+ -> \Sigma^+$. We, hence, study a notion of primality in the endomorphism monoid of the free
semigroup—a natural and fundamental concept in this algebraic structure. We establish that irreducibility is a nontrivial property for the class of so-called rank-preserving endomorphisms, and we provide a characteristic condition separating the reducible and irreducible endomorphisms. We also characterise when an endomorphism is a factor of another endomorphism, analyse the non-uniqueness of factorisations of a rank-preserving endomorphism into its irreducible components, and investigate the use of incidence
matrices to give insights into the (ir-)reducibility of rank-preserving endomorphisms.
Read the full abstract ▾
Machine Learning
SoloQ: Calibration-Free Quantization for Diffusion Language Models
Donghyun Lee, Arkapravo Ghosh, Varun Manjunath et al. (+5 more)
Diffusion large
language models dLLMs) have emerged as a promising alternative to autoregressive
language models through bidirectional
diffusion-based
token generation. However, their growing model sizes and high inference costs make efficient deployment challenging: full-sequence denoising repeatedly invokes compute-intensive forward passes, while block-
diffusion models additionally introduce a
memory-intensive KV-cache. Low-bit weight-activation quantization is therefore attractive, yet existing dLLM post-training quantization methods rely on calibration data despite activation distributions shifting across masking states and denoising steps. We present SoloQ, a calibration-free quantization framework that maps weights and activations into a normalized rotated basis with a predictable marginal distribution, enabling data-independent quantization. SoloQ combines a structured K-RPBH rotation with a lightweight rescaling correction for calibration-free quantization. Its predictable post-rotation distribution supports both distribution-matched codebooks and hardware-native NVFP4. For block-
diffusion models, SoloQ further applies commit-time KV-cache quantization to compress persistent states without perturbing the actively denoised block. Across full-sequence dLLMs (LLaDA and Dream) and block-
diffusion dLLMs(Fast-dLLM v2 and Nemotron-Labs-
Diffusion), SoloQ retains accuracy under 4-bit quantization and outperforms calibration-based baselines on knowledge- and
reasoning-intensive
benchmarks. With NVFP4, SoloQ reduces peak
memory by up to 2.61X and accelerates end-to-end inference by up to 2.24X.
Read the full abstract ▾
All authors Donghyun Lee, Arkapravo Ghosh, Varun Manjunath, Bumjoon Kyle Rhee, Hyunho Kook, Shiting Xiao, Youngeun Kim, Priyadarshini Panda
Human-Computer Interaction
Towards semantic reconstruction of individual words from fnirs using clip loss
Santiago Posso-Murillo, Nathan Palladino, Ben Pyykkonen et al. (+3 more)
Semantic reconstruction maps neural activity to a word-
embedding space, recovering the meaning of a perceived word instead of selecting it from a fixed vocabulary. Functional near-infrared spectroscopy (fNIRS) carries semantic information suitable for this mapping. However, most fNIRS decoders are trained with a squared-error objective that fits each word independently and ignores the geometry of the
embedding space. To address this limitation, we evaluate a contrastive loss based on the contrastive-
language-image-pretraining (CLIP) loss, as an alternative to mean-squared-error (MSE) for reconstructing perceived words from fNIRS. We compare the two objectives by training a bidirectional long short-term
memory (Bi-LSTM) decoder to map fNIRS signals to word
embeddings. We use GloVe-50 and T5 word
embeddings as targets, across three fNIRS
datasets recorded under a shared paradigm pairing each word image with its spoken name. Performance is measured with a pairwise matching score and open-vocabulary top-$k$
retrieval. The Bi-LSTM trained with CLIP is the most consistent decoder across experiments. T5 produces higher matching scores, whereas every significant
retrieval result uses GloVe-50. These results support the use of contrastive objectives as a promising direction for fNIRS semantic decoding and motivate validation on larger
datasets.
Read the full abstract ▾
All authors Santiago Posso-Murillo, Nathan Palladino, Ben Pyykkonen, Dong Y. Han, Luis G. Sanchez-Giraldo, Jihye Bae
Formal Languages and Automata
An Analysis of Decision Problems for Relational Pattern Languages under Various Constraints
Klaus Jansen, Dirk Nowotka, Lis Pirotton et al. (+2 more)
Patterns are words with terminals and variables. The
language of a pattern is the set of words obtained by uniformly substituting all variables with words that contain only terminals. In their original definition, patterns only allow for multiple distinct occurrences of some variables to be related by the equality relation, represented by using the same variable multiple times. In an extended notion, called relational patterns and relational pattern
languages, variables may be related by arbitrary other relations, achieved by using regular patterns and relating individual variables independently from the patterns structure separately. We extend the ongoing investigation of the main decision problems for patterns (namely, the equivalence problem, the inclusion problem, and the membership problem) to relational pattern
languages under a wide range of relevant individual relations, providing a comprehensive foundation in all three research directions.
Read the full abstract ▾
All authors Klaus Jansen, Dirk Nowotka, Lis Pirotton, Corinna Wambsganz, Max Wiedenhöft
Artificial Intelligence
AMBER: Training Long-Horizon Web Agents through Append-Only Memory
Chinmay Savadikar, Zhaoyu Zhang, Mingyu Zhao et al. (+4 more)
Modern
language-model
agents increasingly interact with external environments over long-horizon, multi-step trajectories, where the accumulated interaction history can quickly exceed practical context budgets. To ensure reliability,
agents must maintain factual information over long horizons, remember execution errors and corrective feedback, and track progress across actions. Several approaches have been proposed to achieve this without the need for maintaining the entire execution history in context, such as using the
reasoning and action history, learning to maintain a fixed-size
memory through an overwrite mechanism, and periodic summarization. Although overwrite
memory can in principle retain anything an append-only
memory can, it must learn to carry each fact through every subsequent rewrite, which is difficult to learn from sparse outcome rewards; for interactive applications like web
agents, we find that trained overwrite
memories delete key information required by the trajectory, as well as corrective feedback received from the environment. We introduce AMBER (Append-only
Memory Bank for Evidence Retention) - a simple and scalable framework where an
agent jointly learns to reason, act, and write free-form
memory, while an append-only rule guarantees retention by construction. This allows AMBER to be trained end-to-end with reinforcement learning from outcome rewards without the need for extensive curated SFT data. On WebArena Lite, AMBER improves average success over overwrite-based
memory by 4.09 percentage points, increases the fraction of tasks solved in five repeated runs by 4.8 percentage points, and matches an overwrite baseline trained on substantially more expensive curated supervision. AMBER achieves these improvements while maintaining a practical
token budget, providing a strong balance between context efficiency, task performance, and reliable long-horizon execution.
Read the full abstract ▾
All authors Chinmay Savadikar, Zhaoyu Zhang, Mingyu Zhao, Shuang Xie, Han Li, Tianfu Wu, Lingyun Wang
Computer Vision
R2RI: A Multi-View Event and RGB Dataset for Robot-to-Robot Interaction
Gabriele Magrini, Riccardo Catalini, Federico Becattini et al. (+4 more)
Understanding and modeling interactions between autonomous
agents is a fundamental challenge in robotics, with broad implications for collaborative systems, social robotics, and human-
robot coexistence. Although the study of
robot interactions has emerged as a compelling research direction, progress has been severely hampered by the absence of large-scale
benchmarks. In this paper, we introduce
Robot-to-
Robot Interaction (R2RI), the first
dataset specifically designed to address the
Robot-
Robot Interaction (RRI) task. R2RI consists of different
humanoid robots and realistic interactions modeled on real human social behaviors. Complementary viewpoints are available, i.e., an
egocentric perspective from each
robot's onboard sensors, and an exocentric perspective from external fixed cameras, thus enabling rich spatial and contextual understanding of the interaction dynamics. The
dataset comprises more than $6.5$M frames and $\approx5000$
videos at $120$ fps, including Event and RGB domains. We investigate pros and cons of each domain, comparing state-of-the-art approaches for a number of key sensing and interaction based tasks. We publicly release the
dataset and its annotations for all tasks and modalities at https://github.com/MagriniGabriele/R2RI.
Read the full abstract ▾
All authors Gabriele Magrini, Riccardo Catalini, Federico Becattini, Guido Borghi, Pietro Pala, Roberto Vezzani, Lorenzo Seidenari
Robotics
AIM: Adaptive Interaction Modeling Networks for Real-to-Sim Soft-Body Simulation
Tiancheng Yang, Dingshuo Chen, Tianle Chen et al. (+2 more)
Deformable-object manipulation is essential for robotic tasks such as folding laundry and handling food, where
robots must control shape changes as well as object motion. Predictive soft-body simulation supports these tasks by anticipating deformation under external interactions. However, spatial neighborhoods can misrepresent deformation dependencies, introducing local errors that accumulate over successive predictions. Models fitted to individual scenes must also accommodate changes in object geometry and manipulation conditions. In this work, we propose AIM, an Adaptive Interaction Modeling framework that treats real-to-sim soft-body simulation as a local-global interaction modeling problem. AIM uses motion history and geometry to adapt particle relations over current spatial neighbors and retained connections, while geometry-conditioned global communication coordinates object-wide responses. A unified kinematic control-point interface represents different manipulation configurations, and multi-step autoregressive supervision trains the model on its own predicted trajectories. Experiments on PhysTwin and PGND demonstrate improved motion accuracy and visual fidelity, with a 20.0% reduction in future-prediction tracking error relative to PhysTwin and a 22.8% reduction in mean long-horizon particle error across six object categories relative to PGND. The framework further supports transfer across actions, object instances, and scenes, including zero-shot transfer from
robot interactions to human manipulation without target-domain dynamics fitting.
Read the full abstract ▾
All authors Tiancheng Yang, Dingshuo Chen, Tianle Chen, Zhaocheng Liu, Qiang Liu
Machine Learning
Will the Judge Flip? Predicting Position-Sensitive LLM Judgments from Residual Stream Activations
Hashmath Shaik, Gnaneswar Villuri and Alex Doboli
The order in which candidate responses are presented can change an LLM judge's verdict. Detecting such a position flip ordinarily requires judging each pair in both orders, which doubles the number of judgments. We investigate whether residual stream activations recorded immediately before the initial verdict can predict a flip. We use nested grouped cross-validation to evaluate regularized linear probes on 534 JudgeBench pairs for three Qwen3 judges and Llama-3.1-8B. The linear probes achieve AUROCs of .621-.850 and outperform a combined baseline that uses verbalized confidence, verdict-label logits, response lengths, and the judge's initial choice by .062-.113 AUROC. Linear probes trained on JudgeBench and then frozen achieve AUROCs of .685-.853 on 1,802 MT-Bench comparisons without MT-Bench fitting or recalibration. These results show that pre-verdict activations support prediction of susceptibility to candidate order and outperform the non-activation predictors evaluated here.
Read the full abstract ▾
Machine Learning
Sample-Optimal Estimation of the Fréchet Inception Distance
Ziyun Chen, Jerry Li, Kevin Tian et al. (+1 more)
The Fréchet Inception Distance (FID) is widely used to evaluate generative models, but its empirical plug-in
estimator suffers from finite-sample bias [BSAG18, CF20]. We study the sample complexity $n$ of estimating FID to error $\epsilon$ between $d$-dimensional Gaussians with bounded mean distance and covariances, when one distribution is known. Our contributions are threefold. (1) We establish tight finite-sample $\Theta(\frac{d^2}{n})$ bias and $\Theta(\frac{d}{n} + \frac {d^2} {n^2})$ variance bounds for the empirical plug-in
estimator, establishing a $\gtrsim d^2$ sample complexity. (2) To debias the empirical plug-in
estimator, we generalize the ${\rm FID}_\infty$
estimator of [CF20] to extrapolation methods of arbitrary order $k$. We further prove tight bias and variance bounds of $\Theta(\frac{d^{k + 2}}{n^{k + 1}})$ and $\Theta(\frac d n + \frac{d^2}{n^2})$ for any order-$k$ extrapolation under our framework. (3) We introduce Relative Taylor Debiasing (RTD), a new, computationally efficient FID estimation
algorithm using debiasing techniques inspired by U-statistics. We show that RTD achieves an $O(\frac d {\epsilon^2})$ sample complexity, and prove that this is optimal. We provide a complementary empirical evaluation of our new
estimators. Our experiments on synthetic Gaussians validate the predicted residual bias and support the tightness of our bounds. On ImageNet with Inception
embeddings, RTD achieves the lowest mean estimation error at the standard 50K sample budget, while our second-order variance-aware extrapolation
estimator (VALE$_2$) uses only 10K samples to achieve accuracy comparable to FID$_\infty$ at 50K samples.
Read the full abstract ▾
All authors Ziyun Chen, Jerry Li, Kevin Tian, Yusong Zhu
Machine Learning
LiLib: Lifelong Air-to-Ground Path-Loss Prediction on UAVs via a Drift-Triggered Model Library
Minh Tran
UAVs that act as relays or base stations need accurate air-to-ground path-loss predictions for rate adaptation and placement, but propagation conditions change as a
UAV moves between suburban, urban and high-rise areas, and the same areas are often revisited. Online regressors that adapt by forgetting must relearn each environment from scratch, whereas a single model trained on all data averages incompatible regimes. We propose LiLib, a lightweight continual-learning scheme in which a
UAV maintains a small library of recursive-least-squares experts. A windowed residual test detects drift; a short probe phase then either reuses the best stored expert or creates a new one. In simulations based on four standard urbanization profiles, LiLib reduces prediction RMSE from 5.89 dB (best sliding-window baseline) to 4.03 dB (p < 0.001), lowers the error shortly after a return to a known environment from 12.3 dB to 5.7 dB, and recovers 99% of the throughput of a regime-aware oracle in rate adaptation. The library stores four experts in under 0.5 KB, and identifies regimes with 92% purity without labels. When a second
UAV is initialized with the library of a peer, its error after environment changes halves. LiLib does not reach the oracle, and similar regimes may be merged when shadowing is strong. The results indicate that, for recurring drift, remembering is more effective than re-adapting.
Read the full abstract ▾
Computer Vision
MoonGS: High-quality Representation of the Lunar Surface via Gaussian Splatting Using Robust Depth Features from Image Pairs
Yun Jiang, Bo Zheng, Yingying Zhang et al. (+7 more)
High-quality 3D reconstruction of lunar terrain from sparse rover images is indispensable for autonomous lunar exploration, but remains challenging because viewpoint overlap is insufficient, surface textures are weak, and data volume is limited. We propose MoonGS, the first feed-forward 3D Gaussian Splatting framework tailored to lunar scenes. Given only two input images, MoonGS predicts pixel-aligned Gaussian primitives in a single forward pass and renders photorealistic novel views without any per-scene optimization. MoonGS (i) adopts an adaptable backbone design that seamlessly integrates advanced
vision foundation models to extract robust depth features; (ii) integrates semantic priors in two manners: merging semantic cues with visual features to refine Gaussian parameter estimation, and adopting a semantic ranking loss that regularizes background depth; and (iii) employs an
entropy-guided heuristic resampling strategy to augment sparse observations by selecting the most informative distant viewpoints with negligible overhead. Experiments on the LuSNAR
benchmark and our synthetic weak-texture MoonBlender
dataset show that MoonGS surpasses state-of-the-art feed-forward NeRF/3DGS baselines by +4.9 dB PSNR, +0.29 SSIM, and 40% lower LPIPS while maintaining sub-second inference. Furthermore, we validate the broad applicability of our framework by demonstrating that it effectively leverages state-of-the-art backbones, including VGGT, to significantly boost performance. Qualitative evaluations on Chang'e mission imagery also show the best visual quality among compared methods, indicating robustness on real lunar data. The source code and
dataset are publicly available at https://github.com/InRobots/MoonBlender.
Read the full abstract ▾
All authors Yun Jiang, Bo Zheng, Yingying Zhang, Xueming Xiao, Tao Hu, Hutao Cui, Zhiguo Meng, Ke Gao, Yang Gao, Meibao Yao
Computation and Language
JudgeMoE: Distributional Aggregation for LLM-as-a-Judge
Yiqi Liu, Joseph James, Yang Wang et al. (+3 more)
When an LLM judge scores an output, its score distribution retains uncertainty and disagreement information that is lost after scalar compression. We introduce JudgeMoE, a lightweight aggregator that assigns example-specific weights to cached judge score distributions and fuses them before computing a final score. A protocol study shows that score-range choice is unstable across judge–
dataset settings and that soft scoring usually outperforms hard decoding. On the original 10-
cell benchmark, JudgeMoE improves mean Spearman over uniform log pooling by $+0.079$. Applying the same configuration to six additional
cells yields a $+0.0393$ mean gain over the strongest local single judge across 16
cells, with positive differences in 12/16
cells and a one-sided Wilcoxon signed-rank $p=0.0091$. Validation-based analyses further show that the preferred aggregation method depends on the task and judge pool.
Read the full abstract ▾
All authors Yiqi Liu, Joseph James, Yang Wang, Kun Zhao, Chenghao Xiao, Chenghua Lin
Social and Information Networks
WPTokens: A Dataset for Studying the Regulation of Wikipedia
Léo Joubert
Studying how Wikipedia is regulated requires knowing who adds, rewrites, moves and deletes which piece of content, when and where. WPTokens is a
dataset that answers these questions over the full revision history of the French-
language Wikipedia. It relies on a positional diff over two layers: a text layer made of sentences, and a node layer made of the structural elements of wikitext (templates, links, references, headings, tables, lists, etc.). Each piece of content keeps a stable identifier for its whole life, and every change to it is recorded as an addition, an update, a move, a move with update, or a deletion. Built from the April 2026 dump, the
dataset covers 2.75 million articles, 153.6 million revisions, 342.6 million
tokens and 573.7 million
token events, stored in an indexed SQLite database. A companion R package turns these records into temporal contributor–
token multigraphs, completed by a layer of revert ties between contributors. We describe the tokenization, the diff
algorithm, the data model and the graph construction, and discuss the limitations of the approach.
Read the full abstract ▾
Sound
Neural Representations, Natural Connections: What Transfers From Human Speech Foundation Models to Animal Vocalizations?
Tomás Arias-Vergara, Christopher Hauer, Héloïse Brotier et al. (+3 more)
Speech-pretrained models have shown promise in animal bioacoustics, but the factors governing their cross-species transfer remain poorly understood. We evaluate 15 frozen encoders spanning monolingual and
multilingual speech, speaker verification, animal bioacoustics, and general audio on caller identification across four species and call-type classification across three. Differences between the strongest
speech- and animal-pretrained representations range from $-0.027$ to $+0.119$ UAR;
speech is significantly better in three of seven settings ($p<0.01$) and not significantly different in the remaining four. Controlled comparisons show no systematic advantage from increased
language coverage or animal-domain pretraining, while frozen speaker-verification
embeddings transfer poorly. The results also show that transfer is strongly layer-dependent: raw-waveform models peak early, whereas patch-spectrogram models peak deeper.
Read the full abstract ▾
All authors Tomás Arias-Vergara, Christopher Hauer, Héloïse Brotier, Elmar Nöth, Andreas Maier, Lee Koren
Information Retrieval
Beyond Successor Accuracy: State Retention for Recursive Self-Improvement in Recommendation
Jinfeng Xu, Zheyu Chen, Ziyue Peng et al. (+5 more)
Recommendation recursive self-improvement (Rec-RSI) feeds recommender outputs into subsequent training. Evaluating each round solely through its latest model assumes that the successor consolidates the update, although pre- and post-update models may retain complementary ranking decisions. We term this distributed progress and quantify it using cross-generation advantage (CGA), a marginally matched contrast between cross- and within-generation model pairs. A rank-separation statistic, label-free at selection time, predicts which family to retain. Across four
datasets and three sequential recommendation encoders, the preferred retention regime varies by architecture: cross-generation pairing benefits GRU4Rec and SASRec, whereas FMLP initially favors within-generation pairing and shifts toward cross-generation pairing after a second update. Rank separation selects the stronger family in 12/12 first-update and 5/6 second-update
dataset-encoder settings; on held-out tests, the selected family outperforms the direct successor in 34/36 trajectories. Five transfer mechanisms do not consistently reproduce these gains in one model. These findings establish state retention as a distinct Rec-RSI problem: progress may reside in relations between generations as well as in the latest model. Code is available at \href{https://github.com/Jinfeng-Xu/RecRSI}{https://github.com/Jinfeng-Xu/RecRSI}.
Read the full abstract ▾
All authors Jinfeng Xu, Zheyu Chen, Ziyue Peng, Zheng Lin, Wenhao Yuan, Jian Chen, Shujie Li, Edith Ngai
Game Theory
You All Need to Hear This: Wisdom of Crowds with Deliberating Jurors
Rosa Bauer and Adrian Haret
We analyze a formal model of deliberation in which
agents have probabilistic access to evidence and selectively disclose it to their peers in order to induce opinion change. We present theoretical and simulation results that tackle the question: does deliberation improve the collective accuracy of the group? Through a jury
theorem we chart the conditions that allow wisdom of crowds to emerge after a few steps of deliberation, and, through simulations, we show how this effect is sensitive to the evidence acquisition mechanism and, implicitly, to the diversity in the initial evidence distribution.
Read the full abstract ▾
Machine Learning
Muon Is Theoretically Wrong For Convolutions, But Empirically Effective
Thibaut Boissin, Thomas Massena, Mathieu Serrurier et al. (+1 more)
Muon, an optimizer known for its efficiency, has a clear interpretation for
matrix-valued updates, but convolutional kernels are stored as four-dimensional tensors. Standard implementations reshape these tensors into
matrices, a shortcut which breaks the theoretical understanding behind
Muon. To investigate this, we formalize the corresponding optimization objective directly in convolutional operator geometry and introduce Convolutional Newton-Schulz (Conv-NS), which approximates the polar factor in this geometry while preserving kernel support. When applied in fast training experiments, Conv-NS and reshape-based
Muon are both computationally efficient and achieve comparable accuracy on CIFAR-10 and ImageNet classification tasks. However, as one could expect a theoretically aligned Conv-NS to outperform reshape-based
Muon, we investigate this mismatch between practice and theoretical understanding, with the hypothesis that exact convolutional orthogonalization may overconstrain updates. These findings highlight
Muon's strong practical performance while opening directions for its further development on convolutions. Our code is publicly available at \href{https://github.com/thib-s/muonconv-cifar10-airbench}{github conv-
muon}.
Read the full abstract ▾
All authors Thibaut Boissin, Thomas Massena, Mathieu Serrurier, Franck Mamalet
Networking
Who Can Safely Wait? Risk-Aware 5G Scheduling for Low-Altitude UAV Operation
André Melo, Yueqi Chen, Lourenço Pereira Jr. et al. (+1 more)
Cellular-connected
UAV operations already fly in many countries. However, cellular networks were designed for terrestrial users and remain blind to the operational consequences of delaying information from low-altitude aircraft. We present Risk-Aware (RA), a framework that translates information freshness and operational context into aircraft-specific operational risk, enabling uplink scheduling based on operational consequence rather than network metrics alone. RA balances the consequence of further delay against the service cost of heterogeneous single- and multi-slot updates. It combines a regulation-inspired risk proxy, risk-directed prioritization based on marginal risk growth, and a Max-Weight scheduler with Lyapunov-based performance guarantees. Although RA is derived from an analytically tractable model, we evaluate it under moving
UAV trajectories, dynamic operational context, geographically varying population exposure, and realistic
UAV video traffic. Across 48 operating points and fleets of up to 120
UAVs in São Paulo delivery operations and the Interlagos Formula 1 Grand Prix setting, RA reduces 99th-percentile per-
UAV risk by up to 96% over proportional fair and 86% over round robin, while supporting fleets up to 1.5$\times$ and 3$\times$ larger, respectively, within the same risk envelope.
Read the full abstract ▾
All authors André Melo, Yueqi Chen, Lourenço Pereira Jr., Igor Kadota
Artificial Intelligence
When to Remember, When to Abstain: Category-Conditioned Retention for Reliable Agent Memory
Olukunle Owolabi, Pulkit Gupta and Fei Wang
Persistent
agent memory is only as reliable as its retention decision: an assertion weakly supported by its source can be stored and later reused as established fact. We study whether the retention decision should be governed by a confidence bar conditioned on the semantic category of the assertion rather than by a single global threshold, retaining well-evidenced categories liberally while abstaining more aggressively where inference is unreliable. We evaluate this in a deployed cold-start
memory pipeline on 100 synthetic personas. The empirical evaluation is motivated by a sharp reliability asymmetry: across 4{,}715 candidate assertions, only 77.9% of value and belief assertions are supported by their source, versus 96.2% for all other categories. A global confidence threshold cannot separate these: it either admits unsupported value claims or discards well-evidenced ones. Conditioning the threshold on category resolves the tradeoff. In repeated held-out evaluation, a stricter bar on values alone reduces unsupported retentions from 6.2% to 4.0% (an ${\approx}36\%$ relative reduction, modest but consistent across folds) and, as corroborating evidence, preserves an estimated 13 percentage points more coverage (95% CI 9.8–16.0) than a global threshold at comparable retention. Our results suggest that reliable retention depends on the type of assertion, not on confidence alone, and that a category-conditioned threshold can act as a simple, effective form of selective prediction at the write boundary.
Read the full abstract ▾
Distributed Computing
T-CCL: Resource Efficient and Performant Collective Communication using Tensor Memory Accelerator
Keyvan Dadashzadeh, Yuehong Zhou, Minyu Cui et al. (+1 more)
Large
transformer-based models increasingly depend on multi-GPU execution, which requires frequent collective communication among GPUs. Existing communication libraries often rely on many GPU threads to achieve high bandwidth or low latency, resulting in a large streaming multiprocessor (SM)-side resource footprint. This footprint can limit the resources available to other GPU work, particularly when communication and computation execute concurrently. Thus, efficient collective communication should not only achieve high collective performance but also reduce its SM-side resource usage. This paper presents T-CCL, a resource-efficient collective communication library based on the Tensor
Memory Accelerator (TMA) for intra-node communication. T-CCL offloads both data movement and reduction operations to TMA and executes each collective as a pipelined series of asynchronous TMA operations, reducing the SM resources required for collective communication while maintaining high bandwidth. Evaluated across AllReduce, AllGather, and ReduceScatter collectives, T-CCL outperforms NCCL by up to 2.4x with unrestricted communication resources and up to 3.42x under restricted resource budgets, remains competitive with NCCL's recent symmetric-
memory kernels, and occupies the same or fewer SMs in profiled cases. In a GEMM-collective overlap case study, switching the communication backend from NCCL to T-CCL raises the average operator-level speedup over a sequential baseline from 1.12x to 1.25x on two GPUs and from 1.04x to 1.14x on four GPUs, as T-CCL uses fewer SMs for communication, leaving more SMs available to the overlapped GEMM. Integrated into vLLM as a communication backend, T-CCL improves end-to-end inference throughput over vLLM's automatic backend dispatch by up to 1.31x, outperforming it at every evaluated batch size on both the conversation and decode-heavy workloads.
Read the full abstract ▾
All authors Keyvan Dadashzadeh, Yuehong Zhou, Minyu Cui, Miquel Pericas
Artificial Intelligence
Verified, not generated: expert-verified AI study materials and the distribution of learning gains in a university course
Canh Thien Dang and An Nguyen
Experimental studies of generative AI in education mostly report average effects, yet field evidence shows that AI can narrow attainment gaps or widen them. We argue that the direction depends on the judgement burden, the expertise a learner must supply to screen AI output before learning from it, and that expert verification before release moves this burden from students to an accountable tutor. We test the argument in a two-cohort difference-in-differences design in which one half of a compulsory firstyear university economics course received AI-generated podcasts, FAQs and quiz-based study guides, produced with a source-grounded model and checked by a named graduate teaching assistant (170 students; 340 examination marks). Access was associated with a 2.34-mark advantage on a 50-mark component. The share of marks below the upper-second classification boundary fell by 24.7 percentage points relative to the counterfactual, effects were significant at every threshold from 23 to 31 marks and at none above, and roughly three-quarters of the average originated in the bottom quintile. The threshold estimate is robust to removing the lowest-scoring students from the pre-intervention cohort; the average effect is not. Interviews and feedback from 36 students indicate that the verification label gave students a reason to engage with AI-generated material without ending their scrutiny of it. Evaluations of AI learning resources that report only mean effects cannot detect whether the students the resources are meant to help are the ones who gain.
Read the full abstract ▾
Robotics
TeleHairing: A Teleoperation Baseline for Robotic Haircutting
Zhendai Huang, Aleksi Vilkki, Jianan Huang et al. (+4 more)
Robotic haircutting requires controlled tool motion near the head while simultaneously accounting for communication, visual feedback, tool actuation, and interruption handling. Existing studies still lack an operator-in-the-loop reference for analyzing these coupled behaviors before human trials or stronger autonomy. This paper presents TeleHairing, a closed-loop teleoperation architecture for mannequin-based robotic haircutting evaluation under local, relay, and remote deployment conditions. Logged timing shows that the main remote latency increase occurs before the
robot-side control endpoint: overall timing reached 190.5~ms in remote mode, while
robot-side command queue, control processing, and control-to-
robot timing remained similar across modes. Trajectory analysis shows that the larger remote command-following error was dominated by the terminal withdrawal segment rather than accumulated uniformly over the path; excluding this segment reduced remote root-mean-square error (RMSE) from 45.1 mm to 9.6 mm. Detection-loss trials further show that rebase events resumed motion without a large target jump under the tested condition. These results clarify how deployment, execution, and interruption affect the robotic haircutting teleoperation loop, providing a quantitative reference for future autonomy, safety, and user-facing studies.
Read the full abstract ▾
All authors Zhendai Huang, Aleksi Vilkki, Jianan Huang, Bolin Liao, Chunbo Luo, Leopoldo Angrisani, Shuai Li
Hardware Architecture
Evaluating Inference Compute for Generative AI: A Framework for Enterprise Workloads
Abbas Raza Ali, Muhammad Ajmal Siddiqui and Moona Zahid
LLM deployment is shifting from single-turn completion to agentic trajectories in which a model plans, calls tools, reads results and reasons at test time before acting. This inverts the economics of inference hardware: chat serving amortises weight reads across large batches, whereas
agent trajectories are sequentially dependent, run at effective batch one, and make per-
token decode latency (TPOT) the dominant term in task completion time. Using a roofline analysis and a closed-form episode-latency model, we show why this regime favours accelerators that keep weights in on-die SRAM (Cerebras WSE-3/3T, Groq/NVIDIA LPU) or compiler-managed tiered
memory (SambaNova SN40L/SN50), and why three vendor ecosystems converged in 2026 on disaggregated prefill/decode serving. We show that per-step reliability compounds exponentially in trajectory length-a 2% per-step failure rate erases a 2x decode advantage for a 20-step
agent-so determinism and tail latency are first-order performance variables. We then propose a four-layer evaluation framework (silicon, serving system,
agent episode, enterprise) with a metric set built on goodput at an agentic SLO and cost per successful episode, a six-axis
benchmark protocol over six task families, a paired-bootstrap statistical design, an attestation protocol for vendor-run
benchmarks, and TCO, availability and adoption-timing models with explicit break-even conditions. All performance figures are public and labelled by evidence class; we state seven falsifiable hypotheses and the experiments that test them, and argue that the most likely original result is that
token-throughput rankings diverge from cost-per-successful-task rankings on long-horizon work.
Read the full abstract ▾
Artificial Intelligence
Small Language Models for Smart Data Model Classification at the Edge: A Cost-Aware Hybrid Approach
Cristian Martella, Angelo Martella, Antonella Longo et al. (+1 more)
The rapid proliferation of heterogeneous data sources within the Internet of Things (IoT) across domains such as smart cities, energy management, and environmental monitoring necessitates efficient and scalable data standardization methods. Effective classification of smart data models (SDMs) is essential for facilitating interoperability. However, existing approaches are often limited by high resource consumption and lack applicability in edge environments with constrained computational capabilities. Aiming to bridge this gap, the proposed study evaluates the performance of lightweight open-source
language models (LMs) to resolve an input data entity against its corresponding best fitting SDM representation under resource-constrained conditions. It systematically
benchmarks a diverse array of models, including general purpose (GP),
reasoning-specialized (RS), and code-specialized (CS) architectures, across multiple domain-specific
datasets. Addressing the current omission of lightweight, resource-efficient solutions in the literature, the investigation provides significant and valuable insights into model selection, task formulation, and deployment strategies that optimize accuracy and efficiency. A complementary experiment also compares the surveyed large
language models (LLMs) against two near-zero-cost similarity baselines (Term Frequency-Inverse Document Frequency (TF-IDF) and a lightweight sentence encoder) on the same task, providing a strong reference point for interpreting the practical value of LLM-based classification on edge platforms.
Read the full abstract ▾
All authors Cristian Martella, Angelo Martella, Antonella Longo, Motaz Saad