Papers

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.

The S301 star reaches a pericentre one tenth that of S2, opening a new regime for probing black-hole spin and departures from Schwarzschild. We develop a post-Newtonian secular Hamiltonian framework for bound geodesics in stationary, axisymmetric spacetimes and compute the maximum projected astrometric deviation over one orbit. At the expected GRAVITY+ precision, S301 can probe several nonrotating geometries and sizeable spin-induced deviations, while kilometre-baseline interferometry can probe much smaller departures, establishing S301 as a powerful test of the spacetime geometry of Sgr A*.

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.

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.

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.

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.

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.

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.

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.

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.

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.

Direct-write pattering, using chemical vapor deposition through a nozzle, enables localized film growth using high-purity precursors standard to the electronics industry. Yet, achieving sub 100 $\mu\mathrm{m}$ feature control has been limited by nozzle design and unquantified precursor surface diffusion. Here, we report a direct-write chemical vapor deposition platform that employs a 10 $\mu\mathrm{m}$ tip diameter glass micro-nozzle mounted on a scanning stage above a heated substrate in ultra-high vacuum. Using this approach, we demonstrate the continuous single-pass writing of 500 $\mathrm{nm}$ thick metallic aluminum lines with a 30 $\mu\mathrm{m}$ linewidth on titanium nitride-coated silicon. This capability enables maskless, localized growth under systematically varied deposition conditions on a single substrate. To analyze the resulting deposit profiles, we introduce a transport framework coupling Knudsen gas emission from the micro-nozzle with surface diffusion to extract the effective precursor diffusion length, $L_d$. $L_d$ captures the combined impact of precursor surface mobility, residence time, substrate temperature ($165^{\circ}\mathrm{C}$ to $205^{\circ}\mathrm{C}$), growth duration, and nozzle height, contracting from 17.4 $\mu\mathrm{m}$ down to 3.2 $\mu\mathrm{m}$ at elevated temperatures. These results establish $L_d$ as a quantitative parameter for evaluating laterally confined chemical vapor growth and predicting feature resolution limits. This approach provides a practical methodology to measure and quantify precursor surface transport in localized film synthesis.

A stellar spectrum is determined by a few parameters, so a set of spectra lies on a low-dimensional manifold with those parameters as its coordinates. We introduce Armillary, which generalises the Sequencer of Baron and Ménard from an ordering to multidimensional coordinates, constructed from the spectra alone. Two stars are compared by the Wasserstein distance between their signed absorption profiles. Shortest paths through a nearest-neighbour graph measure distance along the manifold and give the coordinates, and the same graph transfers the labels of a few stars to all the others. The coordinates are more nearly linear in $T_{\rm eff}$ and $\log g$ than principal components, t-SNE, UMAP or an autoencoder. The label transfer, without deep learning, is more precise than the first three and comparable to the autoencoder. On 202,932 APOGEE DR17 stars, an illustrative draw of 32 training stars recovers the ASPCAP $T_{\rm eff}$, $\log g$ and [M/H] of held-out dwarfs and giants, with $1\sigma$ errors of 100 K, 0.18 and 0.083 dex. Resolving the two $\alpha$ sequences of the disc in [$\alpha$/M] for giants takes 128 training stars, and with 2048 the four errors improve to 33 K, 0.06, 0.026 and 0.022 dex. On 104,799 DESI stars, 32 training labels from APOGEE recover the stellar parameters to 129 K, 0.25 and 0.203 dex, even for spectra degraded to a signal-to-noise of about 20. At a signal-to-noise of about 100, 2048 training labels resolve the two $\alpha$ sequences of the disc in DESI giants, with errors against APOGEE of 51 K, 0.07, 0.047 and 0.026 dex. The code is public at https://github.com/tingyuansen/armillary.

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.

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.

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.

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.

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.

Recently, the LHCb Collaboration reported a new narrow bottom-strange meson, $B_{s0}^*(5700)$ with mass $M= (5698.9 \pm 1.5 \pm 0.6)$~MeV, width $\Gamma<9.8$~MeV, and spin-parity $J^P=0^+$ in the $B_s^0\pi^0$ invariant-mass spectrum. Motivated by this observation, we study the low-lying $P$-wave $B_s$ mesons in a two-stage refined constituent quark model. The one-loop radiative corrections to the confinement–one-gluon exchange potential ($\alpha_s^2$) lower the whole multiplet by $25$–$50$~MeV and improve the level ordering; however, the scalar $B_{s0}^*$ mass still exceeds the LHCb data by $\sim120$~MeV in a purely compact picture. The inclusion of the coupled-channel dynamics is decisive: the $BK$ threshold, lying close to the scalar mass, generates a sizeable meson–meson (molecular-like) component that pulls the pole down to $\sim5727$~MeV, within $30$~MeV of the measurement, and simultaneously suppresses the $S$-wave phase space, naturally accounting for the anomalously narrow width. Consistent results are obtained for the $1^+$ $B_{s1}$ partner, for which we predict a mass $\sim5747$–$5755$~MeV, in agreement with the chiral-doubling scenario. The $B_{s1}$ state is expected to be very narrow and decaying mainly through the isospin-violating $B_s^*\pi^0$ and radiative $B_s\gamma$, $B_s^*\gamma$ modes, providing a clear discovery target at LHC Run~3. The $E1$ radiative widths of the $j_\ell=1/2$ states are predicted at the $50\sim 60$~keV, and the strong width of the pure $j_\ell=3/2$ state $B_{s2}^*(5840)$ ($1.5$~MeV) reproduces the measurement. Our results support the interpretation of $B_{s0}^*(5700)$ as a near-threshold scalar with strong molecular dressing and establish the predicted $1^+$ $B_{s1}(5747)$ as the decisive test of the hadronic dynamics in the $B_s$ sector.

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}.

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.

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}.

Kelly's conjecture asks whether every regular tournament admits a Hamilton cycle decomposition. Motivated by its symmetry-preserving extension, we study cyclic Hamilton decompositions of carousel tournaments. For an odd integer $n$, let \[ T_n=\Cay\left( \mathbb Z_n,\left\{1,2,\ldots,\frac{n-1}{2}\right\} \right) \] be the carousel tournament. We ask whether $T_n$ has a Hamilton decomposition invariant under translation by $1$. Although the answer is immediate when $n$ is prime, composite orders introduce a genuine obstruction: nonunit differences generate short cycles rather than Hamilton cycles. We resolve a general composite-order family by proving that, whenever $n=pq$ for primes $7\le p<q$, the tournament $T_n$ admits a cyclic Hamilton cycle decomposition. The proof combines Hamiltonian difference sequences over prime fields with a matching argument that constructs two base paths with disjoint difference sets in $\mathbb Z_{pq}$. Thus our result gives an infinite family supporting the cyclic, symmetry-preserving extension of the Hamilton decomposition problem for regular tournaments.

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.

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.

We prove that every nonconstant Kazhdan–Lusztig polynomial of a sparse paving matroid has only simple negative zeros. For each rank $d\ge5$, the Kazhdan–Lusztig polynomial of the uniform matroid of rank $d-2$ on $d$ elements strictly interlaces those of all sparse paving matroids of rank $d$ and corank at least two.

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.

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.

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.

Although the special theory of relativity was conceived to relate inertial reference frames, it is not generally known by beginning students, or even by academics not specializing in general relativity, that it can be applied to noninertial reference frames. The general perception among non-specialists seems to be that noninertial reference frames can only be described through general relativity. This situation has led to some controversy regarding the necessity of general relativity to address problems involving accelerated systems, such as the twin paradox and the description of rotating reference frames. We discuss the controversy and propose that, to avoid the present confusion, the local validity of special relativity in the absence of gravitation, despite the use of noninertial frames, should be explicitly and clearly stated at an early stage of the learning process.