Statistical Methodology
Dynamic factor and double PCA models for partially observed survival curves: Forecasting demand in short-term rental markets
Marthe Elisabeth Aastveit, Alex Lenkoski and Thordis Thorarinsdottir
This paper develops prediction models for population-level survival curves observed over time and sampled from a heterogeneous mix of populations. We consider a discrete-time setting where each curve is only partially observed and forecasts of the remaining trajectory are needed for downstream decision making. Our approach recasts cross-population heterogeneity into a multivariate sampling model. We propose two
forecasting models for partially observed curves: a full factor analysis model that extends a general factor representation to incorporate the partially observed survival curve, and a double PCA model. The methodology is motivated by demand
forecasting in short-term rental
markets, where
market-level occupancy paths can be viewed as survival curves over the booking horizon and where forecasts of future occupancy feed into dynamic pricing
algorithms. We apply the models to the newly released Wheelhouse
dataset, which contains time series of
market occupancy curves for 500
markets from 2017 to 2022. Model performance is assessed using the integrated quadratic distance, and we compare the proposed PCA-based methods to Holt's linear trend model across multiple forecast horizons. The results show that the proposed models yield accurate and stable forecasts of the remaining survival trajectory and generally outperform Holt's method, particularly at longer horizons.
Read the full abstract ▾
Mathematical Physics
Differential Recursions, Projective Discriminants, and Algebraic Generating Functions
S. Voloshyn
We study two hierarchies of bivariate generating functions defined by differential recursions: a self-dual hierarchy and a reduced-
symmetry hierarchy. Their rational coefficient functions produce palindromic and anti-palindromic triangles, including OEIS A336110 and A140136. For general differential order $p$, the reduced hierarchy decomposes into $p-1$ Horn-type companions. An explicit differential intertwiner of order $p-3$ relates the last companion to the normalized self-dual generating function. The companions share a principal symbol and a rationally parametrized characteristic curve with a projective $S_3$ action. At $p=2$ this construction yields the Narayana quadratic. At $p=3$ the
symmetry lifts to $S_4$: two algebraic sectors are edge resolvents of one quartic root configuration, and the nonsymmetric Gross–Witten–Wadia generating function has degree four. For $p=4$ and $p=5$ we derive algebraic equations of degrees six and eight, respectively, and find Galois groups $S_6$ and $S_8$ at generic specializations. At $p=6$ we verify a degree-ten specialization with Galois group $S_{10}$. Their discriminants exhibit a universal cubic characteristic factor and a squared lower-order factor. A boundary factorization explains the observed algebraic degrees and discriminant monomials. We formulate
conjectures on algebraicity and Galois groups in the reduced hierarchy and identify a possible mechanism for transcendence in the higher self-dual hierarchy.
Read the full abstract ▾
Computers and Society
How can AI accelerate the green transition?
Jacques Sainte-Marie, Marie-Fleur Simmet and Fanny Terrier
In his essay The Gift, published in 1925, the sociologist Marcel Mauss emphasised the 'total' scope of the phenomenon of gift-and counter-gift-giving, noting that it 'expresses, simultaneously and all at once, all manner of institutions: religious, legal and moral (...)\,; economic (...)\,; not to mention the aesthetic phenomena to which these facts give rise (...).” The emergence of artificial intelligence (AI), which is suddenly and profoundly disrupting individual and collective lives, bears a resemblance to the ”total social fact” described a hundred years ago. AI refers to technologies designed to simulate human intelligence, namely an ”automated system that (...) deduces, from received inputs, how to generate output results such as predictions, content, recommendations or decisions that may influence physical or virtual environments.” 4 (OECD). Having become central to human relations, AI relies on physical and globalised infrastructure and therefore, also has an impact on the environment. This paper analyses some of these effects and how they are perceived by society. This text offers an exploration of the links between AI and the ecological transition, seeking to identify new opportunities and constraints. To this end, it will identify the barriers to the ecological transition (1) and the environmental impacts attributable to AI (2). It will then go on to describe how to live in a constrained world (3) while AI accelerates scientific discovery (4) and profoundly transforms society (5).
Read the full abstract ▾
Probability
Universality of cutoff for the Exclusion Process
Jonathan Hermon, Gady Kozma and Justin Salez
Under a fairly general condition on the underlying jump rates, we prove that the Exclusion Process with fixed particle density exhibits cutoff at time $t_{{mix}}\sim \frac{1}{2}{t}_{{rel}} \log N$, where $N$ is the volume and ${t}_{{rel}}$ the relaxation time of the single-particle dynamics. Our result covers, in particular, the standard setting of uniform nearest-neighbor jumps on large discrete
tori in any fixed dimension and, more generally, on any sequence of vertex-transitive graphs with bounded degree and polynomially diverging diameter. Our (short and entirely human) proof combines Wilson's method, the Octopus Inequality,
Fourier analysis on Hamming slices, and a sharp comparison with the Dirichlet form of a natural accelerated variant of the dynamics.
Read the full abstract ▾
Artificial Intelligence
Evolutionary Safety of Recursive Self-Improving AI: Taxonomy, Risk Discovery, and Evaluation
Chang Gong, Jingping Bi, Di Yao et al. (+6 more)
Rapid advances in AI are bringing recursive self-improvement (RSI) closer to practice, with growing progress in model self-training, self-evolving
agents, and automated AI development. Yet the very mechanisms that enable continuous improvement also introduce new challenges for AI safety. Unlike isolated failures, safety risks can persist across updates, accumulate through repeated adaptation, and propagate to successor systems. More importantly, the processes that generate and select future improvements may themselves evolve, making safety increasingly difficult to assess and maintain. This raises a fundamental question, i.e., how can safety be maintained when both AI systems and the processes that improve them continue to change? Therefore, we introduce Evolutionary Safety to study how safety properties are preserved, weakened, or restored throughout persistent and recursive self-improvement. We characterize six recurring manifestations of evolutionary risk and develop a five-domain taxonomy that captures how safety-relevant changes arise, persist, and propagate across evolving systems. Building on this foundation, we establish a framework for risk discovery and evaluation across states, updates, trajectories, and lineages, and examine governance principles for controlling modifications, verifying and authorizing updates, and ensuring traceability and recovery. We further identify open problems in maintaining safety as both AI systems and their improvement mechanisms continue to evolve.
Read the full abstract ▾
All authors Chang Gong, Jingping Bi, Di Yao, Ran Cheng, Yaodong Yang, Kaile Wang, Xinjian Liang, Chao Xiang, Ruijie Guo
Robotics
Onboard Wind-Preview Model Predictive Control Using Pitot-Static Sensing for Multirotor UAVs
Bas Meere, Eline Wisse, Laurens Vousten et al. (+4 more)
Effective wind gust rejection and stable hovering are critical for the outdoor operation of autonomous drones. However, existing gust rejection methods are primarily reactive, inferring the disturbance from the resulting motion or measuring it at the airframe. Either way, the wind has already begun to act before it can be compensated. In this work, we anticipate the gust instead by measuring the wind ahead of the drone with a low-cost, low-weight pitot-static sensor mounted on a boom. The resulting wind preview is incorporated into a nonlinear model predictive controller (MPC), which optimizes the drone motion while anticipating wind disturbances. A longer boom offers more preview time but adds inertia and degrades flight performance. We characterize this trade-off in simulation and show that the optimal preview distance is not a fixed property of the platform, but shifts with the wind speed and with how quickly the drone can respond. Indoor hardware experiments confirm the trend and show that the proposed controller substantially improves hover performance against a PX4 baseline and an otherwise identical wind-unaware MPC. Outdoor experiments show that the error along the wind direction is reduced by 54 percent with respect to the baseline, demonstrating that a single wind-aligned sensor can significantly improve hovering performance.
Read the full abstract ▾
All authors Bas Meere, Eline Wisse, Laurens Vousten, Sander Doodeman, Elena Torta, Paula Chanfreut, Duarte Antunes
Artificial Intelligence
Accounting for Bias Enables Sustainable LLM Evaluation
Harshita Katoch, David Antony Selby, Gerrit Großmann et al. (+1 more)
LLM-as-a-judge has become the de facto standard for scalable, subjective evaluation, yet current leaderboards compensate for systematic measurement bias by running ever more comparisons, an approach that is both statistically unsound and computationally wasteful. The root cause is an incomplete measurement model, treating LLM judges as neutral, interchangeable instruments ignores documented biases like position bias, verbosity bias, judge severity, and self-enhancement, that no volume of additional data can eliminate. We propose a unified latent variable framework that jointly models pairwise and ordinal data while explicitly correcting for these confounders, recovering reliable rankings from substantially fewer comparisons. Because fitting this model costs negligible compute relative to a single round of LLM inference, bias correction is not only more statistically rigorous but also a more sustainable approach to trustworthy evaluation.
Read the full abstract ▾
All authors Harshita Katoch, David Antony Selby, Gerrit Großmann, Sebastian Vollmer
Probability
Recurrence and transience of open Jackson networks with balanced nodes
Serguei Popov
Consider an open Jackson network with Poisson arrivals and exponential services, in which no node is overloaded and exactly~$k_0$ nodes have load equal to~$1$. We prove that the queue-length process is recurrent if and only if $k_0\leq 2$. For that, we show that the network satisfies a sector condition, so its capacities are comparable to those of its symmetrisation; the latter is a reversible network whose recurrence/transience can be dealt with by standard means. We also discuss what the Lyapunov-function method gives in this setting; in particular, we show that every balanced queue is empty at arbitrarily large times, also in the transient case.
Read the full abstract ▾
Numerical Analysis
Energy-minimizing domain decomposition
Muhammad Hassan, Benjamin Stamm and Lambert Theisen
We propose a novel domain decomposition framework based on energy minimization, applicable to a wide class of linear and nonlinear partial differential equations. This energy-minimizing domain decomposition (EMDD) method addresses problems whose solutions are minimizers of an energy functional, either unconstrained or constrained to a
manifold, and thus covers source and
eigenvalue problems in a single formulation. Besides introducing the general framework, we derive the algorithmic realization of the method for four classes of model problems, namely linear and nonlinear source and
eigenvalue problems, and derive the computational complexity of the method. Numerical experiments demonstrate the robustness and efficiency of the method across all four model problems and show that it compares favourably against existing model-specific domain decomposition methods.
Read the full abstract ▾
Computation and Language
Where a Model Sends Its Own Repeated Token
Nicolás Vera Zúñiga
Black-box model identification works by scoring a model's response to natural-
language prompts. One line of work feeds models a degenerate input – their own
token, repeated – to find a failure mode rather than an identity. We take that input and ask where the model goes when it does not. For each
token t, read argmax p(. | t, t) in one forward pass; the result is a map on the whole vocabulary, with two halves. The first – which
tokens are fixed points – is partially anticipated, and we report it as a failed estimand: the natural distance on it is 83% cardinality, separates a corpus manipulation by two bits in 3471 against a precision floor of zero, and attributes families at 0.5833. The second half, where the map sends
tokens that are not fixed points, is unrecorded; the one paper holding those
tokens logged them as a zero. Pairing on the source
token removes the cardinality confound by construction (r from 0.9128 to -0.0932) and attributes families at 0.8333 – twelve models scored against a pool of nineteen – with chance 0.1389, across seven tokenizer groups and several corpora. Two nulls clear it: frequency-matched destinations agree at 0.1429, independent marginals at 0.0798. Family predicts agreement better than tokenizer (0.2031 against 0.1205), and recurrent architectures cluster at balanced accuracy 1.0 against a 0.7895 majority rate, or 0.90 once each model's dominant destination is excluded – the figure we stand behind. We measure the robustness envelope: 8-bit weight rounding moves the map less than deduplicating the training corpus does (0.9004 against 0.6353, on one support), 4-bit destroys it (0.0098; 0.1812 at deployment granularity, so not a coarseness artefact), and the precision floor varies by model from 0.201 to 0.9778. All estimands and kill conditions were registered before the data, and the failed one is reported as fully as the surviving one.
Read the full abstract ▾
Sound
BAT-CLIP: Trimodal Alignment of Brain, Audio and Text
Suhyun Kim, Jinmo Han, Danny Dongyeop Han et al. (+9 more)
Decoding and interpreting naturalistic
speech from the
brain increasingly relies on
alignment to pretrained
speech and
language representation spaces. However, current CLIP-style
brain-
speech alignment ground neural activity to a single anchor modality-audio or text-despite the
brain's inherently
multimodal speech processing. This induces a trade-off: audio anchoring preserves temporal structure but weakens linguistic separability, while text anchoring captures semantics yet discards acoustic detail. We propose BAT-CLIP, the first CLIP-style trimodal
alignment framework for iEEG that jointly aligns neural
embeddings to both pretrained audio and text anchors in a shared, frozen audio-text
manifold. On the naturalistic Podcast
benchmark, BAT-CLIP yields more robust representations than bimodal CLIP baselines. We also highlight the importance of using self-supervised foundation models for CLIP training.
Read the full abstract ▾
All authors Suhyun Kim, Jinmo Han, Danny Dongyeop Han, Ahhyun Lucy Lee, Jewoon Lee, Yonghyeon Gwon, Zach Paris, Chun Kee Chung, Saewoong Bahk, Nam Soo Kim, Seong Jae Hwang, Jiook Cha
Artificial Intelligence
SPO: Discovering Adaptive Large Neighborhood Search Operators via Stackelberg Program Optimization
Xinyi Ke, Kai Li, Junliang Xing et al. (+2 more)
Large neighborhood search (LNS) relies critically on destroy and repair operators, whose effectiveness depends on both adaptation to the evolving LNS state and interaction between the two roles. We introduce Stackelberg Program Optimization (SPO), an LLM-based framework for discovering adaptive executable destroy-repair programs. SPO conditions operator decisions on a compact LNS state, allowing state-dependent behavior to emerge through program discovery, and organizes destroy-repair discovery as a Stackelberg interaction over program space that reflects their asymmetric dependency. Role-specific credits evaluate destroy programs as leaders and repair programs as conditional follower responses, guiding a coupled optimization process that combines LLM generator learning with population-based evolutionary search over programs. Experiments on the traveling salesperson problem and capacitated vehicle routing problem show that SPO outperforms strong baselines across a broad range of settings and generalizes beyond the discovery scale to larger instances and
benchmark sets. Behavioral analyses further demonstrate state-dependent operator behavior and coupled destroy-repair improvement during discovery.
Read the full abstract ▾
All authors Xinyi Ke, Kai Li, Junliang Xing, Yifan Zhang, Jian Cheng
Earth and Planetary Astrophysics
JWST detection of methane emission from a young brown dwarf
K. L. Luhman, C. Alves de Oliveira, J. K. Faherty et al. (+1 more)
In recent studies, we have used low-resolution spectroscopy (R~100) with the James Webb Space
Telescope (JWST) to uncover low-mass brown
dwarfs (2-10 MJup) in the young cluster IC 348 that exhibit the so-called 3.4um absorption band from aliphatic hydrocarbons. Atmospheric models have not predicted the presence of that band in brown
dwarfs at any age. Instead, methane absorption was expected in the coolest newborn brown
dwarfs, but it was not detected. We have performed JWST spectroscopy at higher resolution (R~1000) on one of the brown
dwarfs in IC 348, LRL 11001 (~6 MJup). The new measurement of the 3.4um band is fit well with four Gaussians that represent the CH stretching modes of CH2 and CH3 groups, indicating that the carrier is a large alkane (CnH2n+2). In addition, the spectrum contains emission lines at 3.324 and 3.333um, which we attribute to Q branches of CH4 hot bands between the octad and dyad regions (primarily nu3+nu4->nu4) and the tetradecad and pentad regions. This is the first detection of 3.33um CH4 emission from a young brown
dwarf. The emission probably arises from the photosphere rather than a circumstellar disk given that no disks around
stars or brown
dwarfs are known to exhibit emission in these transitions. Modeling of the emission is needed to constrain the excitation mechanism, which could be either thermal (e.g., an
accretion shock) or pumping by external radiation (e.g., neighboring cluster members). We speculate that vertical mixing has suppressed CH4 and enhanced large alkanes in the photosphere, but CH4 survives in the upper atmosphere, where it is available to experience the observed emission.
Read the full abstract ▾
All authors K. L. Luhman, C. Alves de Oliveira, J. K. Faherty, B. C. Smith
Signal Processing
Floating-Point Microformat Quantization and Pruning for Efficient MU-MIMO Neural Receivers
SaiKrishna Saketh Yellapragada, Esa Ollila, Mário Costa et al. (+1 more)
Neural receivers outperform conventional 5G NR processing chains, but their compute and
memory demands hinder real-time deployment. For a standard-compliant multi-user
MIMO neural receiver, the 4-bit number format, not merely the bit width, determines whether compression preserves the gain over classical receivers. We apply weight and activation quantization-aware training (QAT) and, separately, 50% magnitude
pruning, comparing INT8/INT4 and FP8 (E4M3)/FP4 (E2M1) weights with INT8 post-ReLU activations. Trained on 3GPP UMi channels and evaluated on TDL-B and TDL-C, 8-bit weight-activation models remain within 0.05 dB of FP32 at 10% and 1% block error rate (BLER). At 4 bits, uniform INT4 loses 3.3-3.7 dB and falls below LS-LMMSE, whereas FP4 more than halves this loss (1.3-1.4 dB) and still outperforms it by about 0.5 dB, even after
pruning. FP4's denser near-zero grid matches the trained weight distribution, and FP4 avoids the residual-path over-
pruning seen with INT4. An analytic cost model projects 66x fewer bit-operations and 8.8x less weight storage for pruned 4-bit-weight inference.
Read the full abstract ▾
All authors SaiKrishna Saketh Yellapragada, Esa Ollila, Mário Costa, Yawei Li
Artificial Intelligence
Semantic Navigation for Issue Localization in Code Repository
Yunxiang Wei, Zhenyu Lei and Jundong Li
Repository-level issue localization aims to identify and rank the files and functions relevant to resolving a reported issue. LLM
agents approach this task iteratively: they identify a set of potentially relevant locations, inspect the corresponding code, and revise their judgments about these candidates as new evidence is acquired. Existing environments, however, provide limited support for this loop:
agents must search for unresolved relation targets, reconstruct entity semantics from raw source code, and revise candidates without evidential basis. To address these limitations, we present SemNav, a framework that leverages deterministic
retrieval to seed a broad candidate set and an LLM
agent to continually refine that set, thereby combining initial coverage with evidence-guided revision. SemNav supports this process through three key components. A Semantic
Navigation Graph resolves program relations on demand through a
language server, enabling direct
navigation to related entities across files. Issue-conditioned Semantic Cards provide compact, source-grounded interpretations of each entity's role and relevance to the issue. A persistent Candidate Workspace records each candidate together with its evidential basis, enabling grounded verification, revision, and ranking. Across SWE-bench Lite and PLocBench, SemNav outperforms existing baselines, improving File Hit@10 from 68.33% to 82.67% with Gemma 4B. Component ablations and trajectory analysis support the complementary roles of all three components, while Semantic Cards reduce working-context load by 48.2% relative to full-source reading. SemNav further ranks first on all seven evidence-quality metrics on SWE-Explore and improves downstream issue resolution from 44.00% to 52.33%.
Read the full abstract ▾
Materials Science
Current-voltage characteristics and resistive switching in epitaxial La$_{0.67}$Sr$_{0.33}$MnO$_3$/SrMnO$_3$/La$_{0.67}$Sr$_{0.33}$MnO$_3$ multilayer
A. G. A. Rahman, R. K. Patel, Chandrani Nath et al. (+1 more)
In present study, we investigate the structural properties and current-voltage ($I$-$V$) characteristics of an epitaxial multilayer composed of La$_{0.67}$Sr$_{0.33}$MnO$_3$ (25 nm)/SrMnO$_3$ (70 nm)/La$_{0.67}$Sr$_{0.33}$MnO$_3$ (25 nm), grown on different substrates i.e., SrTiO$_3$(100), LaAlO$_3$(100), Si (100). The used substrates not only have different chemical compositions in line with the film materials, also they have different
lattice parameters which would tune the material chemistry and
lattice strain at the interface significantly. Our $I$-$V$ measurements reveal distinct electrical behaviors depending on the substrate. The multilayers grown on
oxide SrTiO$_3$(100) and LaAlO$_3$(100) substrates exhibit regular $I$-$V$ with slight nonlinearity at low applied voltages. In contrast, $I$-$V$ in multilayer with Si(100) substrate exhibits large asymmetry and notable resistive switching (RS) behavior at room temperature, where a resistance ratio around 10 between the high resistance state (HRS) and low resistance state (LRS) is observed. While the observed $I$-$V$ behavior is sensitive to substrate-induced interfacial disorder and formation of trap bands within bandgap of SrMnO$_3$, the present results have potential applications in future
memory and memristive devices.
Read the full abstract ▾
All authors A. G. A. Rahman, R. K. Patel, Chandrani Nath, A. K. Pramanik
Information Theory
Shift-Invariant Permutations with Polynomial Composition: Construction and Algebraic Structure
Cheng Lyu, Dabin Zheng, Mu Yuan et al. (+1 more)
Shift-invariant permutations of $\mathbb F_2^n$ are particularly attractive in symmetric cryptography because of their regularity and implementation efficiency. Constructing such permutations with an explicit algebraic structure remains a challenging problem. In this paper, we study a family of shift-invariant transformations of $\mathbb F_2^n$ constructed from a recursively defined sequence of mappings $\{\gamma_j\}_{j\ge 0}$ associated with landscapes. First, we prove that, under an explicit dimension condition relating $n$ to the support of the landscape, the sequence $\{\gamma_j\}_{j\ge 0}$ has the
polynomial composition property if and only if the corresponding landscape is quasi-conserved on $\mathbb F_2^n$ with a nonzero special index. Then, we determine the eventual zero or periodic behavior of $\{\gamma_j\}_{j\ge0}$, including the exact index at which the sequence either first vanishes or becomes periodic, together with the least eventual period, and establish linear independence until the first linear relation occurs. When the
polynomial composition property holds, we establish an explicit isomorphism between the group of permutation elements in the monoid generated by these mappings and the unit group of a quotient ring $\mathbb F_2[z]/\langle p(z)\rangle$, where $p(z)$ is a monomial or a binomial determined by the landscape. This framework reduces the tasks of testing bijectivity, computing iterates, and determining compositional inverses and orders of permutation elements to
polynomial computations. Finally, we apply the results to mappings represented by binomials and trinomials in the quotient ring, obtaining explicit criteria and formulas, and tabulate more than one hundred representative constructions of shift-invariant permutations, recovering and unifying several previously studied families.
Read the full abstract ▾
All authors Cheng Lyu, Dabin Zheng, Mu Yuan, Lei Hu
Functional Analysis
Solutions to the 'moment property implies subnormality' and injectivity problems for composition operators on one-circuit graphs
Piotr Budzyński, Zenon Jan Jabłoński, Il Bong Jung et al. (+1 more)
We consider composition operators in $L^2(\mu)$, where $\mu$ is a discrete measure. Our main result shows that, for every integer $p\ge2$, there exists a non-hyponormal composition operator that generates Stieltjes moment sequences and whose symbol induces a directed graph consisting of a
circuit of length $p$ and one infinite branch. In contrast, if the symbol induces a directed graph consisting of one loop and one infinite branch, then the generation of Stieltjes moment sequences does imply subnormality. These two results completely solve the “moment property implies subnormality” problem left open in [Adv. Math. 310 (2017), 484-556] for the entire family of one-branch directed graphs with a
circuit of arbitrary length. Combining them with the examples in that paper shows that, for directed graphs consisting of one loop and a finite number of infinite branches, $2$ is the smallest number of branches for which the moment property does not imply subnormality. Finally, we construct a noninjective composition operator that generates Stieltjes moment sequences and whose symbol induces a directed graph consisting of one loop, one infinite branch, and one leaf. This operator gives a negative answer to the injectivity problem posed in the aforementioned paper. Both constructions use the Krein and Friedrichs measures of an arbitrary normalized S-indeterminate Stieltjes moment sequence.
Read the full abstract ▾
All authors Piotr Budzyński, Zenon Jan Jabłoński, Il Bong Jung, Jan Stochel
Systems and Control
Data-Driven Switched State Estimation with Sparse Sensor Scheduling for Nonlinear Networked Systems
Gianfranco Gagliardi, Franco Angelo Torchiaro, Vincenzo Gallelli et al. (+2 more)
This paper presents a data-driven framework for joint observer design and sparse sensor scheduling for unknown nonlinear networked systems. The nonlinear dynamics are approximated online as a piecewise sequence of locally linearized discrete-time systems, resulting in a switched linear representation recursively identified through Subspace State-Space System Identification (4SID). To ensure consistency across regime transitions, an Orthogonal Procrustes
alignment is introduced to promote coordinate consistency across consecutive regime transitions and mitigate artificial discontinuities caused by arbitrary state-space coordinate changes. Based on the identified local realizations, a dual-rate predictor-corrector observer is designed. The observer gain is computed through a convex optimization problem that jointly addresses estimation accuracy, sensor sparsity, and stability requirements. In particular, an L_{2,1}-norm regularization term promotes column sparsity in the gain
matrix, enabling the automatic selection of informative measurement channels, while a spectral-norm constraint guarantees Schur stability of the estimation error dynamics. The proposed framework is first validated on a traffic network simulated in Aimsun Next. Results obtained on an 18-link network show that the observer accurately reconstructs macroscopic traffic states while significantly reducing the number of active sensors required for real-time estimation.
Read the full abstract ▾
All authors Gianfranco Gagliardi, Franco Angelo Torchiaro, Vincenzo Gallelli, Ayman El Qemmah, Alessandro Casavola
Materials Science
Modern triad for magnetic material science: geometric shape, electronic bands and spin textures
Denys Makarov, Oleksandr Pylypovskyi, Rui Xu et al. (+1 more)
To enable novel concepts in nanotechnologies, it is insightful to design materials with qualitatively new performances originating from recent discoveries in fundamental material science and physics. Curvilinear magnetism emerged as a tool to design
chiral and anisotropic responses of materials at the nanoscale using the effects of the
geometric shape and
topology of the object. This concept is distinct yet complementary to material screening, which is primarily based on the optimization of intrinsic microscopic properties to modify
magnetic textures for specific applications. Contemporary curvilinear magnetism focuses on mean-field micromagnetics, while the explicit contribution of the
lattice structure and electronic degrees of freedom remains beyond consideration. In this review, we will emphasize on the effects which appear when this coupling is accounted for and that have resulted in novel concepts for the design of “metageometric” materials. Further development of this field will provide appealing technological prospects for fundamental understanding of curvilinear
magnetic nanostructures and their applications in energy efficient and scalable nanoelectronics.
Read the full abstract ▾
All authors Denys Makarov, Oleksandr Pylypovskyi, Rui Xu, Carmine Ortix
Computer Vision
TaskIR: Task-Driven Image Restoration via Degradation Adaptation and Task Feedback
Yanjie Tu, Qingsen Yan, Axi Niu et al. (+4 more)
Task-driven image restoration aims to improve both image quality and downstream task performance. However, existing methods predominantly focus on single degradation type and struggle to handle the diverse degradations encountered in real-world scenarios. Different degradations impose distinct restoration demands, and insufficient restoration may leave residual degradations and artifacts that impair object boundaries and semantic cues, thereby compromising downstream task performance. To address these challenges, we propose TaskIR, a two-stage task-driven unified image restoration framework that integrates degradation-adaptive restoration with task feedback refinement. In Stage I, a Degradation Representation Module (DRM) extracts degradation representations, enabling a Degradation-Guided
Transformer Block (DGTB) to dynamically modulate feature transformations for adaptive restoration. In Stage II, a Task-to-Restoration Feedback Generation module (TRFG) transforms heterogeneous task features into restoration feedback by modeling task-representation discrepancies associated with the current restoration. Subsequently, a Selective Task Feedback Refinement module (STFR) assesses feedback relevance and selectively refines intermediate restoration features to mitigate interference with well-restored content. Extensive experiments demonstrate that TaskIR achieves competitive restoration quality and downstream task performance across diverse degradations and tasks.
Read the full abstract ▾
All authors Yanjie Tu, Qingsen Yan, Axi Niu, Wenxuan Cai, Tao Hu, Wei Dong, Haokui Zhang
Computation and Language
Improving Visual Sensitivity of LLMs on Multimodal Machine Translation with Metric-based Loss Weighting
Paweł Mąka, Piotr Andruszkiewicz, Yusuf Can Semerci et al. (+2 more)
Multimodal Machine Translation aims to incorporate additional signal from non-textual modalities to improve translations by resolving ambiguities. While models, through
multimodal fusion, are able to accept images related to the source text, they can ignore this information. Therefore, increasing their visual sensitivity remains an active research area. In this work, we introduce a training method, Metric-based Loss Weighting, that improves visual grounding of translations by increasing the loss function for
tokens that benefit from the accompanying image. We identify these
tokens using the Point-wise Cross-mutual Information (PCXMI) metric, which compares the model's output probabilities with and without visual context. We introduce a Congruency-based PCXMI metric and experimentally show that both metrics working in combination yield the best results. We evaluate our method by fine-tuning three pretrained
Multimodal Large
Language Models on the task of Image-guided Machine Translation for three
language directions. Metric-based Loss Weighting outperforms other tested methods on the CoMMuTE contrastive
dataset, improving accuracy by up to more than 7 percentage points compared to standard fine-tuning, while maintaining strong general translation performance.
Read the full abstract ▾
All authors Paweł Mąka, Piotr Andruszkiewicz, Yusuf Can Semerci, Jan Scholtes, Gerasimos Spanakis
Machine Learning
Audio emotion recognition for atypical hearing
Ulysse Roussel
My doctoral work aims to explore Audio
Emotion Recognition (AER) in the context of atypical listening. This research focuses on auditory hypersensitivity in people with autism, a phenomenon that is often difficult to evaluate and unique to each individual. Our core idea is to leverage our understanding of affect from acoustic traits, relying on the possibility of generalizing affective responses from a small amount of annotated data. As a first step, we fine-tune a large foundation model, Contrastive
Language-Audio Pretraining (CLAP) using low-rank adaptation (LoRA), trained on a valence and arousal
dataset of neurotypical listeners.
Read the full abstract ▾
Artificial Intelligence
Neural State Prediction: Obstructing Shortcut Learning in EEG Foundation Models
Kieren Yu, Ziyang Liu, Chang Huang et al. (+2 more)
EEG foundation models increasingly use masked prediction to learn from unlabeled recordings, but optimizing this objective does not ensure transferable neural representations. A central challenge is that stable positional cues and local correlations can make masked regions predictable without integrating distributed neural context. To reduce this reliance on low-information prediction paths, we introduce Neural State Prediction (NSP), a latent-predictive framework that constrains both the prediction target and the available context. NSP uses a Target Encoder updated by an exponential moving average (EMA) to define latent supervision. Identity residualization removes additive effects associated with channel identity and relative time from the targets, while
topology-separated context excludes their immediate spatial and temporal neighborhood from the visible input. We pretrain NSP on 2.2 million
EEG segments from TUEG and evaluate it across 30 downstream
datasets spanning clinical diagnosis, sleep staging,
emotion recognition, motor imagery, event-related potentials, cognitive-state decoding, and
language retrieval. Under full-parameter multi-task fine-tuning on
EEG-FM-Bench, NSP achieves 63.94 macro balanced accuracy across 14
datasets, exceeding the strongest evaluated baseline by 2.35 percentage points. Controlled component ablations assess the contribution of each mechanism, while matched context controls and held-out interventions characterize the role of context geometry, signal content, and positional information. Jointly designing latent targets and their context offers a promising direction for
EEG foundation models that learn from distributed signal structure.
Read the full abstract ▾
All authors Kieren Yu, Ziyang Liu, Chang Huang, Jintai Chen, Kaishun Wu
Information Retrieval
AgentRecommender: LLM Agents Enable Customizable Recommender Systems on the User Side
Ryoma Sato
Recommender systems have traditionally been developed for platforms. However, this has given rise to many phenomena that may be advantageous for platform lock-in but are a nuisance to users, such as clickbait, filter bubbles, and the spread of fake news. Recently, user-side recommender systems have been proposed as a new paradigm for solving this problem. If users deploy their own recommender systems, they are no longer at the mercy of the platform's interests. However, building a user-side recommender system is not trivial; in particular, customizing one for oneself requires additional data. We propose AgentRecommender, a method that leverages the investigation capability and internal knowledge of LLM
agents to flexibly build user-side recommender systems without additional data. AgentRecommender allows users to easily create recommender systems tailored to their own preferences.
Read the full abstract ▾
Sound
BreathGRU: A Novel Semi-Supervised Bidirectional Gated Recurrent Unit Framework for Speech and Breath Segmentation for Respiratory Audio
Sania Fatima Sayed, John W. Holloway, Reyer Zwiggelaar et al. (+1 more)
Speech-breath
segmentation is a fundamental preprocessing step in respiratory audio analysis, enabling applications such as respiratory acoustic biomarker extraction, lung function prediction and disease monitoring. Existing approaches, including threshold methods,
Fourier Transform-based techniques, and unsupervised and pretrained voice activity detection (VAD) models, primarily focus on
speech detection and often classify breathing events as non-
speech or silence, limiting their applicability for precise breath detection. To address this limitation, we propose BreathGRU, a semi-supervised Bidirectional Gated Recurrent Unit (BiGRU) framework specifically designed for
speech-breath
segmentation. The proposed framework combines frame-level acoustic feature extraction with bidirectional recurrent modelling, pseudo-label refinement and duration-constrained Segmental Viterbi decoding to produce
speech and breath
segmentation. BreathGRU was evaluated against the existing approaches, using manually annotated recordings. Performance was assessed using event-based, time-based, overlap-based, duration-based and boundary-based
segmentation metrics. Experiment results demonstrated that BreathGRU achieved the highest breath event recall (0.83), the lowest onset-localisation error (0.14s) and the highest Mean Match Intersection over Union (0.81), with competitive overall
segmentation performance compared to large pretrained VAD models like Silero. Qualitative evaluation on manually annotated recordings further showed close agreement between BreathGRU and manual annotation, with better breath detection compared to Silero. These findings demonstrate that explicit breath event modelling provides advantages over general-purpose VAD models and establish BreathGRU as an effective
speech-breath
segmentation framework which can be applied for respiratory audio analysis and pulmonary healthcare applications.
Read the full abstract ▾
All authors Sania Fatima Sayed, John W. Holloway, Reyer Zwiggelaar, Faisal I. Rezwan
Information Retrieval
SPADE: Escaping the Popularity-Similarity Frontier to Measure Serendipitous Recommendations
Tobias Vente, Maarten Peirsman, Noah Daniëls et al. (+2 more)
Recommender systems engineer serendipity to foster active exploration and break predictable consumption cycles. The problem with existing offline beyond-accuracy metrics is that they often either isolate historical similarity or global popularity. We aim to design an evaluation metric that examines similarity, popularity, and actual user relevance. To achieve this, we introduce SPADE (Serendipitous Pareto Distance Evaluation). SPADE maps all items into a two-dimensional space to directly calculate a user-specific Pareto frontier of maximally popular and historically similar items. The final serendipity score is then computed by averaging the minimum Euclidean distance from this boundary strictly for the correctly recommended test-set items. Evaluating SPADE across five
datasets and five baseline
algorithms confirms its effectiveness; our results show that the metric successfully prevents
algorithms from exploiting beyond-accuracy measures with irrelevant or non-personalized recommendations, reliably isolating serendipitous discoveries.
Read the full abstract ▾
All authors Tobias Vente, Maarten Peirsman, Noah Daniëls, Hannu Toivonen, Bart Goethals
Optics
Probing temperature dependent non-radiative decay channels in CrSBr with nonlinear optics
Minjiang Dan, Till Weickhardt, Paul Herrmann et al. (+9 more)
Among the recently discovered layered
magnetic materials, the antiferromagnet CrSBr has attracted considerable interest thanks to its stability, quasi-one-dimensional
crystal structure, and direct near-infrared band gap. These properties make it a suitable platform for combined electrical,
magnetic, and optical measurements, and thus for studying
exciton dynamics and transport across different
magnetic phases. In this work, we investigate the optical response of CrSBr across its paramagnetic and
antiferromagnetic phases by combining nonlinear optical spectroscopy, including second-harmonic generation, third-harmonic generation, and two-
photon photoluminescence, with linear photoluminescence measurements. The observation of a clear difference in the temperature dependence of third-harmonic generation and photoluminescence intensities allows us to identify the onset of non-radiative recombination channels in the
antiferromagnetic phase.
Read the full abstract ▾
All authors Minjiang Dan, Till Weickhardt, Paul Herrmann, Fabian Glatz, Marie-Christin Heißenbüttel, Thorsten Deilmann, Michael Rohlfing, Ksenia Mosina, Zdeněk Sofer, Benjamin Pingault, Julian Klein, Giancarlo Soavi
Machine Learning
Yuhan Zhu, Xiangfei Qiu, Hanyin Cheng et al. (+5 more)
Time series
forecasting is typically framed as learning a direct mapping from historical to future observations in the observation space. However, sequences of observations generally provide only a partial view of the dynamics of the underlying system, with future observations being shaped by latent dynamics. Recent latent-space
forecasting methods thus achieve improved performance by predicting future observations from latent-space representations of historical observations rather than directly
forecasting future observations in the observation space. Next, while future observations are also shaped by external factors, how to incorporate external, often
multimodal, information into
forecasting, so that it can shape latent-state formation and
evolution directly, remains underexplored. We propose WorldTS, a world-modeling based
forecasting framework that integrates
multimodal covariates directly into the
forecasting to further improve
forecasting performance. Specifically, WorldTS employs a two-stage training strategy. First, it learns
forecasting-relevant latent state dynamics conditioned on
multimodal covariates, yielding encoded future states. Next, the learned state dynamics are frozen, and an observation decoder is trained to map the predicted future states back to future observations. Extensive experiments on 21 real-world
datasets offer insight into WorldTS and its effectiveness.
Read the full abstract ▾
All authors Yuhan Zhu, Xiangfei Qiu, Hanyin Cheng, Wangmeng Shen, Chenjuan Guo, Bin Yang, Jilin Hu, Christian S. Jensen
Machine Learning
I Act Therefore I Am: When Is JEPA's Action-Conditioning Enough to Learn Causal Mechanisms?
Yuhang Liu, Zhuo Huang and Javen Qinfeng Shi
Recent empirical and theoretical advances suggest that joint-
embedding predictive architectures (JEPAs) may learn meaningful representations for action-conditioned prediction of future outcomes, thus becoming one of the foundational structures for world models. However, accurate prediction does not, in general, necessarily imply recovery of underlying causal states that give rise to the observed dynamics. This work investigates when and how JEPAs can recover the underlying causal states from observations. We first introduce a latent variable model, in which high-dimensional observations are generated from latent causal states whose dynamics are governed by action-conditioned transition mechanisms. Based on this formulation, we develop a general information-theoretic objective that combines conditional likelihood maximization for learning transition dynamics with
entropy maximization for preserving latent state information. We then establish identifiability conditions under which representations learned by this general objective recover the underlying latent causal states up to component-wise invertible transformations and permutation. One key condition for such identifiability is sufficient action-induced variation in the transition mechanisms. Guided by this finding, we instantiate the general objective with an action-modulated Gaussian additive-noise model, yielding action-modulated JEPA (A-JEPA). Experiments on synthetic environments verify the theoretical findings under the identifiability conditions and robustness to moderate violations, while visual
benchmarks demonstrate improved state recovery and transfer to unseen transition mechanisms.
Read the full abstract ▾