Foundation models offer a promising route to compress multi-modal physiological signals into compact representations of human health, with broad applications across sleep medicine, cardiology, neurology and other healthcare domains. Existing models have typically been trained with masked-reconstruction or contrastive objectives. However, masked reconstruction may be poorly suited to the stochastic nature of these signals, while contrastive approaches rely on positive-pair definitions despite the semantic invariances of physiological signals being poorly understood. In this work, we show that next-token prediction is a simple and scalable alternative. We develop Hypnos, a multi-modal sleep foundation model trained using eight different sensing modalities (e.g. EEG, ECG, respiratory signals) drawn from over 20{,}000 overnight polysomnography recordings. We tokenize each modality into streams of discrete tokens using residual vector quantization, then train a large auto-regressive RQ-Transformer to jointly predict the next token across all modalities in parallel. After training, Hypnos can be applied to continuous streams of sensor data from any subset of supported modalities, generating embeddings for downstream tasks. Hypnos significantly outperforms existing foundation models on most benchmarks evaluated. In sleep stage classification, we match the performance of strong supervised baselines on held-out test sets whilst using 100\times less labelled data. Hypnos even generalises beyond sleep recordings, surpassing a dedicated ECG foundation model at detecting atrial fibrillation. Our results demonstrate that auto-regressive token modelling is a strong self-supervised approach for representation learning from multi-modal physiological signals. Code and model weights: https://github.com/joncarter1/hypnos
Predicting beyond the training domain necessarily requires assumptions about how the ground truth extrapolates, yet in most neural networks these assumptions are made only implicit through the architecture. We propose TrustMix, a lightweight post-hoc method that makes a pretrained model's extrapolative behaviour explicit and controllable. TrustMix combines the predictive distribution produced by a pretrained model with a user-specified fallback distribution $p_0(y|x)$, such that as the out-of-distribution (OOD) deviation statistic grows, we collapse towards $p_0(y|x)$. TrustMix provides graceful degradation: the predictive distribution converges toward the fallback distribution as the OOD signal grows. The fallback distribution $p_0(y|x)$ can reflect domain knowledge, if available, or rely on a domain-agnostic stationarity assumption. TrustMix does not require retraining of the base model and can be applied to both regression and classification problems. We evaluated TrustMix on several synthetic and real-world distribution shifts, including tabular, image, and natural language processing tasks. We tested TrustMix on datasets such as CIFAR-10/100 and ImageNet and on numerous architectures including Convolutional Neural Networks (CNNs) and Transformers. TrustMix remains competitive in-distribution while improving predictive uncertainty under distribution shift, with gains in predictive likelihood for regression and calibration for classification. Finally, we showed that TrustMix can provide useful predictive distributions for decision-making applications by utilizing it as a surrogate for Bayesian Optimisation.
Kaden Uhlig, Kiarash Mokhtari, Anıl Arslan, Yunsu Kim, Joern Wuebker
Frontier evaluations of tool-using conversational agents are conducted almost exclusively in English, leaving a critical gap for organizations deploying agents in other languages and regions. We present $τ$-bench-LILT, a multilingual extension of ${τ^3}$-bench covering German, Korean, Turkish, and Farsi in the airline, retail, telecom, and banking domains. We natively localize every layer of the benchmark: customer databases are populated with culturally appropriate records, user-simulator prompts are refined iteratively through contrastive annotation by native-language experts, and tasks are updated according to the localized database and to reflect realistic tasks in the target region. Evaluation across eight agent models shows the largest drops compared to English in the Korean and Farsi retail domains, by up to 61 points in pass^4. Code and data released at https://github.com/lilt/tau-bench-lilt.
Abdullah Al Noman, Gaurav Singh, Rameez Ahmed, Debashri Roy
Grant-free uplink access in low-Earth-orbit (LEO) non-terrestrial networks (NTNs) poses a coordination challenge when multiple UEs share a satellite channel without real-time centralized scheduling under long propagation delays and time-varying channels. Conventional random access does not exploit peer context, leading to collisions and inefficient channel use as contention increases. To address this problem, we propose IRIS, a multi-agent reinforcement learning framework for decentralized grant-free direct-to-satellite (D2S) uplink access, with A2A-IRIS as its sidelink-assisted Agent-to-Agent (A2A) realization. Each UE executes a shared binary transmission policy using local traffic/channel information augmented by compact peer states exchanged over UE-to-UE sidelink, while Vanilla-IRIS serves as a local-only ablation. Evaluations in rural and suburban LEO D2S deployments use TLE-based trajectories propagated using the Simplified General Perturbations 4 (SGP4), with a Walker-Delta constellation as fallback. A2A-IRIS achieves 2.1579 and 5.4478 Mbps, corresponding to 99.9% and 96.1% of the respective scenario-specific upper bounds. Relative to Vanilla-IRIS, throughput improves by 40.8% and 325.0%, with collision rates of 0.0000 and 0.0433, while one- and two-epoch delayed centralized schedulers are outperformed by approximately 51--130%. A2A-IRIS also improves fairness, service coverage, and transmission efficiency with moderate Deep Q-Network (DQN) inference overhead. Active-UE load scalability results further show sustained performance over the evaluated load range. These results demonstrate the effectiveness of sidelink-assisted A2A coordination for decentralized LEO D2S uplink access under propagation delay and delayed channel feedback.
Each new GPU generation is showcased by its peak instruction rates, yet users hardly ever run instructions directly: they reach the hardware's performance through vendor libraries, applications and devices shared with other jobs. How much of a theoretical speedup survives these layers is rarely measured, so a team deciding whether to move to a new generation, or to adopt FP8 or 2:4 sparsity, lacks adequate guidance. AMD's MI300A (CDNA3) adds new FP8 matrix instructions and 2:4 structured-sparse instructions that the predecessor, the MI250X (CDNA2), does not have. We analyze this theoretical speedup gap on AMD's MI300A (CDNA3) against the MI250X (CDNA2) by defining a new metric, capability gain. Capability gain is the fraction of a feature's instruction-level speedup that it keeps at a higher level, such as a library, an application or a shared device. We trace it from ISA microbenchmarks through hipBLASLt and hipSPARSELt to LLM inference and mixed-precision HPL, and onto a device shared by up to eight concurrent streams. FP8 keeps most of its speedup through the library, completing an $8192^3$ GEMM \genCapPkgEndToEndGainEightK{} faster than the MI250X with FP16, but unfused quantization turns it into a per-layer slowdown in an LLM, and 2:4 sparsity keeps at most about half of its speedup in hipSPARSELt. Under queue sharing, FP8's GEMM capability gain rises above 1. How fairly streams share the device depends on the number of hardware queues, not on the numeric precision, yet shared LLM serving gains nothing from FP8.
We introduce LinAlg-Bench, a diagnostic benchmark evaluating 10 frontier large language models on structured linear algebra computation across a strict dimensional gradient of 3x3, 4x4, and 5x5 matrices. Spanning 9 task types and 660 SymPy-certified problems, the benchmark exhaustively evaluates 6,600 model outputs. Beyond binary accuracy, LinAlg-Bench introduces a three-stage automated forensic pipeline classifying 1,156 failures into ten primary error tags with fine-grained subtypes, revealing that LLM mathematical failure is not random but structurally constrained by algorithm type and matrix dimension. Our central finding is a sharp behavioral threshold at 4x4 scale: below it, models fail through execution errors -- sign tracking failures, arithmetic drift, and parity errors; above it, failure transitions to computational abandonment, with models fabricating responses through tool roleplay, constraint-consistent confabulation, and structured hallucination rather than attempting computation. This fabrication-to-abandonment transition is near-universal across all model tiers and architectures, suggesting a working memory limit rather than a knowledge gap, supported by three scale-emergent error types absent at 3x3 but present at 4x4 and 5x5. We further show that solution strategy rigidity is a near-perfect predictor of 5x5 determinant accuracy, document constraint-aware confabulation as a novel structured hallucination failure mode, and release all data, model outputs, error labels, and judge pipeline publicly.
Álvaro F. Silva, Martim Afonso, Nuno Saavedra, Emily First, Alexandra Mendes, João F. Ferreira
Coding agents achieve high success rates on existing benchmarks for Verus, a verifier for Rust programs. These benchmarks typically provide the auxiliary lemmas needed to complete a proof. In practice, however, such lemmas must be discovered and proved. To capture this challenge, we present SEVERUS, a context-aware technique to generate Verus benchmarks that organizes verification tasks into quests, each divided into levelsin a gamified manner. Using the original proof dependencies in verified projects, SEVERUS constructs one quest per verification target and progressively removes proof context to create individual levels. These levels range from the easiest, requiring completing only a single missing lemma, to the hardest, requiring proving the target without any supporting lemmas. An independent checker rejects changes to executable code, weakened specifications, and unsound proof constructs. A noisy binary search estimates the highest level that an agent can solve reliably. Applied to six open-source Verus projects, SEVERUS generates 819 quests comprising 10,095 levels. Experiments with Codex using GPT-6 Sol and GPT-6 Luna, and with Claude Code using Sonnet 4.6, show that success decreases and cost increases as proof context is withdrawn. The strongest agent solves the final level of 52 of 60 sampled quests, yet can only reliably solve 35% of their levels, with failures concentrated in longer quests. Successful GPT-6 Sol proofs rarely reproduce the withheld lemmas, instead, using alternative proof decompositions. By systematically varying proof context, SEVERUS provides a foundation for constructing verification benchmarks that can continue to challenge increasingly capable coding agents.
Mingyang Jiang, Yamin Li, Fan Ma, Daniel Moyer, Hua Xu, Catie Chang
Electroencephalography (EEG) foundation models learn representations that combine information across channels and time. This structure motivates adapters that model the same relationships during downstream learning. However, it is not clear whether such a design outperforms other candidate adapters. Moreover, in order to arrive at a fair comparison between candidate adapters, we also need to consider their trainable parameters. Here, we ask whether a backbone-informed (native) adapter makes better use of a given capacity by comparing it against a token-wise multilayer perceptron control that transforms each feature vector independently. We approximately match their trainable parameter counts and keep the pretrained encoder fixed. Across two EEG foundation models and tasks covering emotion recognition, sleep staging, and EEG event classification, the native adapter shows no consistent advantage. For one backbone, differences between the designs are small relative to variation across runs. For the other, the token-wise control performs better on emotion recognition and EEG event classification. On sleep staging, the native adapter's advantage over the token-wise control is reduced when we also train the component that processes sequences of EEG segments. The relative performance of the two adapters depends on the backbone, task, and downstream training protocol. These findings motivate evaluating backbone-informed adapters against capacity-matched alternatives.
Low-Earth-Orbit (LEO) satellite networks are reshaping maritime connectivity, offering much lower latency and higher capacity than traditional geostationary (GEO) links. As a single user terminal (UT) cannot serve hundreds to thousands of passengers, a cruise ship often deploys several UTs behind an onboard multi-WAN router for aggregated capacity and load balancing. Consequently, passenger connectivity depends not only on the satellite links but also on how the router distributes sessions among UTs. In this paper, we measure Starlink maritime connectivity on two cruise ships of different scales, Utopia of the Seas near the Bahamas in December 2025 and Silver Muse in the Adriatic Sea in July 2026, combining UT telemetry, inside-out measurements from a passenger device, and outside-in measurements from cloud servers near Starlink Points of Presence (PoPs). We find that the onboard network, rather than the satellite links, largely determines the passenger experience, as cruise lines rate limit each passenger device. Moreover, the public IP address of a passenger device changes not only at PoP handovers but also within the same PoP. Furthermore, while the country code reported by the UT reflects the maritime jurisdiction of the ship, the serving PoP follows neither this jurisdiction nor geographic proximity alone, suggesting factors beyond the distance-based PoP boundaries observed in Starlink aviation. To the best of our knowledge, this is the first systematic study of Starlink on commercial cruise ships, complementing prior land mobile and in-flight measurements.
Generative retrieval (GR) has emerged as a strong candidate generator for sequential recommendation as language models and semantic-item identifiers have improved. However, GR is not a uniformly stronger replacement of dense retrieval (DR) in online retail candidate generation. We examine their differences on two Amazon Toys datasets under different metrics. GR leads on best relevant item value among users both methods cover, while DR leads on exact-target recall and relevant value across the ranked list. With deep analysis, we find that popularity contributes the most to GR's measured value lead. Adding the same static item value to DR in one reranking step reverses this lead and keeps recall above GR. Finally, on an industrial-scale internal e-commerce dataset with LLM-judged relevance, we show that a 14B SID-LLM still underperforms a $47\times$ smaller dense encoder.
Radiologists read a screening mammogram by first forming a global impression of the four views and then selectively magnifying suspicious regions before assigning a BI-RADS assessment. MLLMs, in contrast, see the whole image at a fixed, reduced resolution, which makes sub-millimeter calcifications invisible, and they perform close to chance on mammography. Therefore, we present MammoSeek, a mammography agent that reproduces this reading process. After forming an impression of each breast, the agent performs selective zooming: it decides by itself which detector candidates deserve a closer look and inspects only those on full-resolution views. Guided by a BI-RADS assessment skill, it then assigns a category and generates a localized, lexicon-structured report for each breast; its assessment is combined with image-model scores in a decision layer to decide recall. On an internal (EMBED) and an external (VinDr-Mammo) dataset, \ours reaches exam-level AUCs of 0.824 and 0.818, far above other MLLMs, mammography agent and CNN classifiers. Selective zooming also outperforms both no zooming and forced zooming in localization, structured-report F1 and recall decisions at low cost.
Reasoning is essential for autonomous driving (AD) in long-tail scenarios, where vehicles must apply commonsense knowledge, understand spatial relations, infer agent interactions, and make safe decisions. However, existing AD datasets and benchmarks mainly target perception, prediction, or planning, and provide limited supervision for reasoning over realistic long-tail driving scenes. We introduce nuReasoning, a large-scale real-world dataset and benchmark for reasoning-centric AD. Following the lineage of nuScenes and nuPlan, nuReasoning advances real-world AD datasets and benchmarks toward reasoning in long-tail driving scenarios. The dataset contains 20,000 clips, each 20 seconds long, collected across multiple cities, with synchronized multi-camera images, LiDAR data, HD maps, object annotations, and human-verified reasoning annotations spanning Spatial Reasoning, Decision Reasoning, and Counterfactual Reasoning. Unlike prior datasets that focus primarily on visual question answering, nuReasoning supports both reasoning evaluation and planning evaluation, enabling a direct study of how reasoning supervision affects driving performance. Experiments show that fine-tuning VLMs on nuReasoning substantially improves driving-specific question answering, while incorporating reasoning supervision into VLA training improves planning performance even when textual reasoning outputs are disabled at inference time. These results establish nuReasoning as a foundation for evaluating and improving robust, interpretable, reasoning-driven AD systems in realistic long-tail settings.
Rohan Subramanian Thomas, Shikhar Shiromani, Abdullah Chaudhry, Ruizhe Li, Vasu Sharma, Kevin Zhu
Prompt design affects the moral judgments and refusal behavior produced by large language models (LLMs), yet controlled comparisons remain fragmented across datasets and models. We introduce ProMoral-Bench, a benchmark-style comparison of 10 prompting paradigms across four LLM families. Using ETHICS, Scruples, WildJailbreak, and our 200-pair ETHICS-Contrast set, we report task-level metrics together with Moral Competence Score (MCS), jailbreak/over-refusal Safety Robustness Score (SRS), and Unified Moral Safety Score (UMSS). UMSS is a study-specific relative summary of MCS and SRS, not an absolute measure of model safety or moral grounding. In this fixed-template evaluation, compact exemplar-guided scaffolds attain stronger competence-robustness trade-offs at lower token cost than the particular multi-stage variants tested, although multi-stage prompting occasionally improves jailbreak refusal. The few-shot demonstrations are non-overlapping but drawn from the evaluation distributions, so their advantage should not be read as out-of-distribution robustness. Code and data are available at https://github.com/rohanthomas334-sketch/ProMoral_Bench.
Pruning network connections is key to brain function but, despite its importance, there exist few biologically-plausible pruning rules with demonstrated good performance. We evaluate noise-prune, a recently introduced unsupervised local pruning rule for recurrent networks that uses noisy fluctuations to determine the importance of connections. Noise-prune has previously only been empirically tested on random networks without a specific computational function. We show that noise-prune preserves task-performance in task-trained recurrent neural networks, greatly outperforming a strategy that only uses connection magnitude and exceeding a non-local strategy that uses second-order information. Rather than deterministically removing connections that fall below a threshold, noise-prune samples connections to preserve based on their importance and strengthens retained connections to preserve average strength. We show that this sampling and rescaling is essential to good performance, and can augment other pruning strategies. We also find that the optimal empirical degree of rescaling is lower than predicted by the original theoretical argument, and thus develop a new version of noise-prune that outperforms the previous version. Our results demonstrate that a biologically-motivated pruning strategy based on local activity statistics can effectively preserve task performance in functional recurrent networks by combining activity-dependent connection selection with compensatory strengthening of surviving connections.
Pietro Barbiero, Giovanni De Felice, Mateo Espinosa Zarlenga, Francesco Giannini, Filippo Bonchi, Mateja Jamnik, Giuseppe Marra, Ruggero Noris
As Artificial Intelligence models grow in complexity, interpretability has become an indispensable tool for understanding, debugging, and controlling their computations. However, interpretability lacks general theories to deductively design interpretable methods. This gap between theories and methods results in a fragmented literature and inconsistent evaluation protocols. To fill this gap, we introduce the Standard Interpretable Model (SIM), a general theory that enables the deductive design of interpretable methods from first principles. Specifically, the SIM summarises, in a set of premises, what interpretability is for a target entity. From these premises, the SIM systematically derives interpretability symmetries and corresponding constraints. One can then build interpretable models by either introducing constraints in the optimisation problem or by compiling constraints into an interpretable architecture. The most important finding, both theoretical and empirical, is that full expressivity and interpretability symmetries can coexist in the same model, showing that the trade-off between accuracy and interpretability is an artefact of model design. We demonstrate this on language models (LMs), where SIM-based LMs match the performance of equivalent opaque LMs without making any compromise on interpretability. Beyond this, the SIM shows how to improve the interpretability of existing pre-trained models, highlights underexplored research directions, and unifies a discipline that has long been fragmented, offering pedagogical grounding for interpretability curricula.
Amirhossein Abaskohi, Giuseppe Carenini, Peter West, Yuhang He
Large language models increasingly operate over long contexts, where the KV cache grows linearly with sequence length and dominates memory during decoding. Token eviction discards entries permanently, and semantic compression fixes each unit's resolution at prefill, so detail that becomes relevant later cannot be recovered. We propose SeKV, a resolution-adaptive KV cache built around a trained zoom-in mechanism that decides at decode time which compressed spans to restore to token resolution. SeKV segments the context at high-surprisal positions, keeps anchor tokens and span summaries on GPU, and offloads low-rank factors of each span's keys and values to CPU. Lightweight per-head routers select query-relevant spans, which are reconstructed with a learned per-span rank. The cache is lossy but non-evicting: every span stays in attention, coarse or reconstructed. The base LLM stays frozen, and the router adds under 0.1% parameters. Across five backbones and four benchmarks at a 10% GPU-resident KV budget, SeKV outperforms every method that fixes resolution at prefill, improving over the strongest semantic baseline by 2.9 points (5.5% relative). It matches exact-recall offloading caches on LongBench and InfiniteBench and trails on synthetic exact retrieval, while transferring 3 times less data per decoding step than ShadowKV and using the least total memory among the methods we profile. At 128K context, it halves peak GPU memory and cuts total memory by one third relative to full KV caching.
We argue that AI slop in software development is creating a tragedy of the commons. Slop is cheap to generate but expensive to review, and its costs fall on shared resources such as reviewer capacity, which was limited even before AI assistants became widespread. Ostrom documented how communities avoid such tragedies by setting and enforcing their own rules. We translate Ostrom's design principles for sustainable commons institutions into concrete proposals for contributors, reviewers, tool developers, team leads, organizational leaders, and educators.
Łukasz Struski, Hanna Blazhko, Piotr Kubaty, Jacek Tabor
Fast linear algebra in deep learning usually comes with a choice: fixed geometry and exact computation, as in the Fourier transform, or adaptive geometry paid for by dense parameters, random features, or low-rank surrogates. To move beyond this trade-off, we introduce LAPLEX, a class of exact, trainable (phased) Laplace-kernel operators. A LAPLEX layer is a typically full-rank dense matrix, implicitly defined by learnable coordinate anchors, with FFT-like scaling. Consequently, it supports trainable matrix--vector operations at vector dimensions up to $10^9$ on modern GPUs. As a neural layer, it yields compact projections and classification heads interpretable as soft, trainable routing models. The same primitive also serves as an efficient Gram operator, enabling high-dimensional covariance models on flattened images of dimension $3 \cdot 10^6$ that preserve visible spatial structure without imposing convolutional bias. These applications reflect a single principle: dense geometry can be learned without storing a dense matrix, which enables data-adaptive global interactions in regimes where ordinary dense layers are out of reach. In this sense, LAPLEX separates expressivity from storage cost: it behaves like a dense trainable matrix, but is represented and applied through a small structured set of parameters.
Looped transformers promise test-time compute scaling by spending more iterations on harder problems, but it remains unclear which architectural choices let them extrapolate to harder problems at test time rather than memorize training-specific solutions. We introduce a fixed-point based framework for analyzing looped architectures along three axes of stability -- local attractivity, input-dependence, and geometry -- and use it to characterize when fixed-point iteration yields meaningful predictions. Theoretically, we prove that looped networks without recall have countable fixed points and cannot achieve strong input-dependence at any spectral regime, while recall with outer normalization reliably produces a regime in which fixed points are simultaneously locally-attractive, locally smooth in the input, and supported by stable backpropagation. Empirically, we train small looped transformers on chess, sudoku, and prefix-sums and find that downstream performance tracks the framework's predictions across tasks and architectural configurations. We additionally introduce internal recall, a novel recall placement variant allowing an identity residual path, and exploit its contrasting fixed-point behavior to test our framework's predictions in a controlled setting.
This paper treats chord progression generation as a standalone task and asks how much old-domain data must be retained in fine-tuning to acquire a new genre without forgetting the old. I study jazz fine-tuning of a pop-pretrained 25M-parameter Music Transformer (84.21% top-1 next-token accuracy on a held-out pop test set). The jazz corpus is over two orders of magnitude smaller than the pop corpus, so every run uses all 1,513 jazz training sequences, and the swept variable is the volume of pop "rehearsal" data mixed alongside: 0, 1K, 2.5K, 5K, or 10K. At the tabulated best epochs every fine-tuned run gains 7 to 9 points of jazz top-1 (the released F1 checkpoint equals the Phase 0 baseline, as the paper's Correction explains), and pop top-1 falls by 2.11 points under jazz-only fine-tuning but by at most 1.19 with rehearsal. Top-1 also scores the three header tokens of each sequence, and two of them shift whole runs: the jazz-majority runs predict the jazz genre marker even for pop songs, and most test songs carry a placeholder C-major key that a checkpoint predicts for every such song or for none, as the best epochs of F1 and F3 happen to do. Scored without the header tokens, every run gains 7.2 to 7.8 jazz points, jazz-only fine-tuning loses 1.48 pop points, and the four rehearsal volumes lose pop at nearly one rate, within 0.11 points of each other at every epoch and 0.73 to 0.90 points below baseline by their last epoch. The smallest volume tested (0.66x the jazz volume) already gives nearly all the measured protection, and no volume stops a slow decline. In informal listening the author more often selects the 10K and 1K endpoints, which carry more committed stylistic identities, than the middle of the sweep, but no formal listening study has been conducted. All six checkpoints and the selection-corrected F1 retrain ft-pop80-v2 are released at https://huggingface.co/PearlLeeStudio.
Adrian Kruse, Kadir Yilmaz, Daan de Geus, Francis Engelmann, Bastian Leibe
Self-supervised learning has given rise to vision foundation models that learn general-purpose image representations transferable across downstream tasks, yet representation learning for 3D scenes has not achieved comparable success. Motivated by this, we identify key limitations of current 3D representation learning approaches and revisit their core design choices. We first replace the commonly used hierarchical PTv3 backbone with the Volume Transformer, a ViT-like architecture for 3D scenes. We then redesign two core objectives of modern approaches built around student-teacher networks: matching representations across views and predicting masked content from visible context. For the first objective, we separate the roles of teacher and student: the teacher provides clean, consistent targets while the student learns invariance to stronger perturbations. For the second objective, we prevent the student from accessing the masked geometry and instead require it to infer missing representations from visible context only. Finally, we redesign 2D-to-3D knowledge distillation, replacing direct regression of view-dependent image features with view-consistent supervision. Together, these changes form VINO, a framework that learns semantic representations of 3D scenes without labels. VINO substantially improves dense linear probing across indoor and outdoor benchmarks and transfers strongly to out-of-domain settings.
Sign languages are natural, visual-gestural languages used by Deaf communities worldwide. Over 300 distinct sign languages remain severely low-resource due to limited documentation, sparse datasets, and insufficient computational tools. This systematic review synthesizes literature on sign language recognition and translation for under-resourced languages, using Azerbaijan Sign Language (AzSL) as a case study. Analysis of global initiatives extracts eight actionable lessons, including community co-design, dialectal diversity capture, and privacy-preserving pose-based representations. Turkic sign languages (Kazakh, Turkish, Azerbaijani) receive special attention, as linguistic proximity enables effective transfer learning. We propose three paradigm shifts: from architecture-centric to data-centric AI, from signer-independent to signer-adaptive systems, and from reference-based to task-specific evaluation metrics. A technical roadmap for AzSL leverages lightweight MediaPipe-based architectures, community-validated annotations, and offline-first deployment. Progress requires sustained interdisciplinary collaboration centered on Deaf communities to ensure cultural authenticity, ethical governance, and practical communication benefit.
MapHub, the MIDAS arXiv Preprint Hub, assesses each day's new arXiv submissions for every participant in a research community, with a large language model reading each paper against portfolios built for the participants. It publishes the relevance scores, from 0 to 100, as a heatmap that shows whose work each paper touches. In the authors' blind ratings, the top 10\% of papers by MapHub's scores held about 70\% of the raters' own top 5\%, pooled, against about a third for a text-matching baseline. About 6\% of a day's astro-ph papers reach 30 for each participant, and one every three days reaches 60. MapHub runs on a standard Claude subscription and is designed to be useful for any field: https://github.com/kailicao/maphub-public.
Distributed Quantum Computing (DQC) offers a promising pathway to scale superconducting quantum hardware, but restricted inter- and intra-module connectivity creates a major transpilation bottleneck. Current DQC transpilation methods often assume idealized hardware connectivity, ignore subsequent intra-module routing, or suffer from substantial computational overheads that limit their applicability at scale. To address this, we introduce a scalable transpilation method tailored for the restricted connectivity of distributed superconducting quantum architectures. Specifically, our quantum interaction graph (QIG) approach couples a graph-partitioning mapping step with a fast SABRE-inspired routing heuristic. On a 48-qubit (3 QPUs $\times$ 16 qubits) architecture with all-to-all inter-core connectivity, our method reduces average aggregated costs by up to 24.72% (unstructured) and 24.01% (structured), while delivering >250$\times$ speed-ups for structured workloads compared to the state-of-the-art DMapS algorithm. On the 399-qubit (3$\times$133) IBM Flamingo, where DMapS times out on dense circuits, our approach maintains practical runtimes. Against a DQC-adapted SABRE baseline, it reduces structured circuit routing costs by up to 51.94% (all-to-all inter-core topology) and 51.96% (restricted line topology). Ultimately, this work delivers a practical transpilation method for DQC at scale.
This report treats chord-symbol sequences as an interpretable, controllable time series for genre-local harmonic modeling. The frozen Music Transformer base is weight-identical to a pop-dominated Phase-0 baseline. I adapt it to eleven target genres: blues, bossa nova, Bach chorales, country, electronic, folk, funk, gospel, hip-hop, R&B/soul, and rock. The main grid compares LoRA, IA3, BitFit, prefix tuning, and full fine-tuning over 11 genres and 3 seeds (165 cells). All five methods improve over the frozen base on held-out chord prediction (macro gains +2.89 to +3.61 percentage points). Top-1 also scores the three header tokens of each sequence, which move whole runs: only a matched adapter predicts its genre marker, and whether a run predicts the placeholder C-major key of every Chordonomicon song is close to arbitrary. Without the header tokens the macro gains are +1.64 to +2.38 pp, and outside Bach chorales LoRA, IA3, prefix tuning, and full fine-tuning give the same gain (+1.30 to +1.34 pp). That gain reflects a handicap of the frozen base, which for every target but rock reads a genre marker it never read in training: given the pop marker instead, it comes within 0.12 pp of the released matched adapter in every non-chorale genre. Wrong-genre adapters beat the frozen base only where it reads such a marker, and outside Bach chorales the matched adapter's edge over the best wrong one is within 0.2 pp. Only on Bach chorales does adaptation add clearly more (+9.6 pp for LoRA over the base given the pop marker). Further diagnostics (rank sweeps, matched-size and same-weights controls, chord-only classification, real songs) support a bounded conclusion: for genres the base already covers, chord-symbol adaptation mainly supplies a usable genre marker, and chord symbols alone do not carry complete genre identity. Perceived genre authenticity is left to controlled listener evaluation.
David Rundel, R. Teal Witter, Santo M. A. R. Thies, Matthias Feurer, Fabian Fumagalli, Bernd Bischl
Shapley values are a central tool for attribution in machine learning. Since their exact computation requires evaluating the cooperative game to be explained on exponentially many player coalitions, they are typically estimated from a limited number of evaluations. Recently, ShaplEIG has shown that adaptively selecting the coalitions to evaluate substantially improves sample efficiency over classical estimators. This, however, comes at the cost of a sequential procedure that, in every iteration, refits a Gaussian process surrogate of the game and selects the coalition with the largest expected information gain about the Shapley values. The resulting computational overhead limits its applicability to costly games with low evaluation budgets. We present fast adaptive Shapley value estimation with Odd Gaussian Processes (ShapleyOG) to overcome this limitation. ShapleyOG leverages the fact that Shapley values depend only on the odd component of the game, and we show that under paired sampling a Gaussian process surrogate on this component alone yields the same Shapley value posterior and coalition selection criterion from half as many observations. Moreover, ShapleyOG refits the surrogate only on a sparse schedule and efficiently updates the selection criterion in between, and it extends quadrature-based computation schemes to the prior covariance of the Shapley values, which allows for exact or cheaper approximate computation. In extensive experiments on diverse games, ShapleyOG matches the estimation accuracy of ShaplEIG while reducing its overhead by up to three orders of magnitude. As a result, it outperforms state-of-the-art non-adaptive estimators at equal total computation time already when a single evaluation takes a tenth of a second and a budget of a few thousand evaluations suffices, thereby widening the regime in which adaptive coalition selection is applicable and pays off.
Inverse optimization (IO) seeks to infer the parameters of a decision-maker's objective from observed context--action data. We study noiseless IO, where demonstrations are generated by a ground-truth objective. We provide a high-probability ${O}(\frac{d}{T})$ generalization bound for the induced action set, where $d$ is the number of unknown parameters and $T$ is the size of the training dataset. We strengthen this guarantee under additional conditions that ensure uniqueness of the chosen action, bringing our IO guarantees in line with best-arm identification results in the bandit literature. We further show that the ${O}(\frac{d}{T})$ rate is tight over all consistent estimators considered here, and extend the result to both instantaneous and cumulative regret. Notably, the resulting regret lower bound matches the corresponding upper bounds in the adversarial setting, indicating that the stochastic IO setting is effectively adversarial for the class of estimators studied here. Finally, we propose a parameter-free algorithm with lower per-iteration complexity than generic solvers. Experiments validate the predicted rates and illustrate the tightness of our bounds.
Self-evolving LLM agents have gained prominence for their ability to improve after deployment by modifying their harness, including their controller instructions, memory management protocols, and reusable tools and skills, in response to user and environment feedback. However, locally useful updates may persist into later tasks where they produce unsafe behavior, even without direct adversarial influence. To study this risk, we introduce SEABench, a benchmark for studying endogenous misalignment arising from agent self-evolution, with 48 longitudinal task sequences that span multiple evolution surfaces, task domains, and harm types in a rich personal-assistant environment. To account for the stochasticity inherent in agentic operations, we provide an adaptive trajectory discovery pipeline that probes for failures while preserving original task intent and supports causal attribution through paired non-evolving agents and attribution scores. Our evaluation across multiple recent LLMs, evolution surfaces, and harm types reveals that self-evolution indeed increases task completion rates but often at the cost of safety failures that are absent for paired non-evolving baseline agents. We also show that qualitatively different safety behaviors emerge across evolution surfaces and harm types. Further, we show that this divergence in safety behavior is reflected in agents' chain-of-thought reasoning, which yields an effective monitoring strategy that can mitigate unsafe behavior with a low false positive rate.
As Large Language Models (LLMs) move from curated training sets into open-ended real-world environments, a fundamental limitation emerges: static training cannot keep pace with continual deployment environment change. Scaling training-time and inference-time compute improves static capability but does not close this train-deploy gap. We argue that addressing this limitation requires a new scaling axis-evolution. Existing deployment-time adaptation methods, whether parametric fine-tuning or heuristic memory accumulation, lack the strategic agency needed to diagnose failures and produce durable improvements. Our position is that agentic evolution represents the inevitable future of LLM adaptation, elevating evolution itself from a fixed pipeline to an autonomous evolver agent. We instantiate this vision in a general framework, A-Evolve, which treats deployment-time improvement as a deliberate, goal-directed optimization process over persistent system state. We further propose the evolution-scaling hypothesis: the capacity for adaptation scales with the compute allocated to evolution, positioning agentic evolution as a scalable path toward sustained, open-ended adaptation in the real world. Our code is publicly available at https://github.com/A-EVO-Lab/a-evolve.
Global surveillance systems record large volumes of network traffic for retrospective analysis, but packet captures alone reveal little about protocol-level semantics. This paper presents an automated, specification-guided methodology for reconstructing passive Dolev-Yao intruder executions from captured network traffic and validating the reconstructed executions against a formal protocol specification. The framework takes a PCAP capture, an AnBx specification, and the associated deployment metadata; it extracts Java-serialised objects from TCP streams, reconstructs protocol sessions and channel-to-step assignments, and generates a passive intruder that evaluates the Dolev-Yao derived checks produced by the AnBx compiler. Three reconstruction challenges are addressed: multiple serialised objects and object references within a TCP stream, ambiguity in mapping interleaved connections to concurrent protocol sessions, and semantic validation of candidate reconstructions. We evaluate the methodology on 121 protocols from the AnBx test suite. For single-session captures, the standard JDK cryptographic provider recovers 97.8% of expected objects and achieves 94.2% successful reconstruction, where success requires complete object recovery and a clean passive-intruder execution. For the 112 parallel-capable protocols, the toolchain achieves 89.3% successful reconstruction for both two and three concurrent sessions, while recovering approximately 97.2% of expected objects. Experiments with mismatched protocol-role mappings and deliberately mutated traces show that the generated Dolev-Yao checks respond to semantic inconsistencies. By making the reconstruction step explicit, the solution produces protocol-level observations on which assessments of protocol resilience and the design of defensive countermeasures can be built.