Combinatorics
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Friday 6 November
We prove that every IC-planar graph has weak-strict degeneracy at most $6$. Weak-strict degeneracy is the weak degeneracy of Bernshteyn and Lee plus one, so it bounds the coloring parameters controlled by weak degeneracy. In particular, every IC-planar graph is $6$-choosable, $6$-DP-colorable and $6$-paintable, which strengthens the theorem of Yang, Wang, Wang and Lih that IC-planar graphs are $6$-choosable. An IC-planar graph is a subgraph of a plane graph in which a set of pairwise vertex-disjoint inner $4$-faces, called marked faces, receive both diagonals. Our main tool is a Thomassen-type statement for such graphs: vertices of marked faces get one extra unit, except those in a prescribed independent set of boundary vertices. With no marked faces, this statement is exactly the theorem of Bernshteyn, Lee and Smith-Roberge that gives weak-strict degeneracy at most $5$ for planar graphs, so it contains both results. We prove it by a minimal-counterexample argument following Bernshteyn, Lee and Smith-Roberge. The new ingredients are a deletion rule that turns destroyed marked faces into ordinary edges, a pseudo-$2$-chord lemma for paths that use one diagonal, and a choice of orientation of the outer cycle that avoids a vertex of the independent set joined to two vertices of the prescribed boundary path.
Robotics
Toward the Era of Agentic Embodied Intelligence
Foundation models increasingly guide robot actions, yet sustained autonomy requires efficient control and failure recovery. We introduce A2-MEAS, an embodied multi-agent framework built around code as a scaling interface. Learned correction codes combine corrective language, spatial evidence, execution history, and action proposals. MetaBrain provides training supervision and runtime guidance. A correction encoder and adapter absorb this knowledge into policy weights during training, while fresh codes adapt behavior with frozen weights at deployment. The action model, A2-WLA, adds motion-aware history and a latent world model to a vision-language-action policy. Surprisingly, this simple design with action pretraining and task specialization reaches 0.5755 Q-score on the 50-task BEHAVIOR 2026 public subset before online correction, 2.21x the 2025 winner's reported 0.2605 across different benchmark editions and training configurations. On all 100 BEHAVIOR 2026 public tasks, A2-MEAS achieves 0.6411 Q-score and 45.40% task success. On RoboTwin 2.0, A2-WLA achieves 93.98% average success, rising to 95.05% with the complete system. Across 12 real-robot tasks in five household settings, A2-WLA achieves 57.92% average success versus 26.67% for pi0.5. In long-horizon pour-over coffee preparation, a history-conditioned adapter raises post-failure completion at two filter-handling steps from 14% to 94% and from 8% to 82%, compared with A2-WLA retries. Our training infrastructure delivers 427.1 samples per second, 4.03x the throughput of a pi0.5 reference implementation on matched hardware. We will release the system, models, training scripts, and infrastructure to support efficient embodied learning.
Read the full abstract
Computer Vision
Look Closer, Align Better: Explicit Visual Attributes for Robust EEG-to-Image Decoding
EEG-to-image retrieval commonly aligns neural responses with representations from pretrained vision encoders. However, these representations may underrepresent fine-grained visual details needed to distinguish semantically similar images. To address this limitation, we first investigate the EEG decodability and complementary predictive value of explicit visual attributes, identifying visual cues that capture EEG-relevant information beyond pretrained representations. Building on these findings, we propose AttrEcho, a framework that enriches pretrained visual representations with complementary, spatially aligned visual attributes. Specifically, Region-Aligned Visual Attribute Fusion (RAVF) integrates EEG-relevant local visual attributes with spatially corresponding pretrained features through a learnable signed residual gate. We further employ Feature Token Aggregation (FTA), which learns dimension-specific weights across regions to aggregate refined visual features for contrastive alignment with EEG representations. Experiments on THINGS-EEG demonstrate that AttrEcho achieves 94.8% Top-1 and 99.7% Top-5 accuracy in intra-subject 200-way retrieval, while maintaining 82.1% Top-1 accuracy as the gallery expands to 1,600 images with same-concept distractors. AttrEcho also achieves 25.3% Top-1 and 53.0% Top-5 accuracy in inter-subject retrieval. These results highlight the importance of EEG-relevant visual attributes and spatially aligned fusion for fine-grained EEG-to-image retrieval.
Read the full abstract
Sound
Improving Large Audio-Language Model Embedding Baselines for Instruction-Aware Speech Retrieval
Speech retrieval finds relevant spoken documents given a spoken query. The spoken query alone may not reveal the retrieval goal, such as finding an answer or matching speaker identity. Instruction-aware speech retrieval expresses this goal through a natural-language text instruction. Large audio-language models (LALMs) effectively process speech and text, and our analysis shows that different retrieval requirements favor hidden states from different audio-encoder layers. We introduce IRIS-Embed, which adapts LALMs into embedding models through contrastive training and instruction-aware audio layer weighting. We construct about 900K training samples from real data and synthetic data, with instructions expressing requirements for one or more speech attributes. On INSPIRE, an instruction-aware speech retrieval benchmark, IRIS-Embed consistently outperforms the corresponding LALM baselines. The best model achieves 62.92% macro-average Recall@50, surpassing the strongest baseline at 45.15%. Controlled comparisons demonstrate instruction awareness and the benefits of audio layer weighting. On MAEB, a massive audio embedding benchmark, macro-average gains of 9.4-19.8 points over the corresponding baselines show generalization beyond speech retrieval.
Read the full abstract
Computer Vision
CoMET: Tracking the Shared Pose Error of Roadside Messages for Asynchronous Vehicle-Infrastructure Cooperative 3D Detection
A roadside sensor can detect vehicles hidden from the ego vehicle, but a pose error shifts every detection in the same message. We study asynchronous vehicle-infrastructure cooperative 3D detection on V2X-Seq using only information available at decision time, without ground truth, future frames, or per-pair calibration offsets at inference. Objects seen by both sensors reveal the shared translation, but even at zero added delay, 38% of validation messages have no matched pair. We propose CoMET (Common-Mode Error Tracking), which estimates this translation over time. At each message arrival, CoMET matches vehicle and roadside detections, estimates the offset from mutually nearest pairs, and updates a Kalman filter for that sender. When no pair matches, the filter carries its estimate forward. The correction applies to all roadside boxes, including those unseen by the ego vehicle. The estimator has no learned parameters; a separate frozen head ranks the merged detections. Across four delays from 0 to 400 ms on 21 validation sequences, CoMET averages 66.63 bird's-eye-view AP at IoU 0.7, compared with 63.18 for the same late fusion using uncorrected online poses and 62.96 for the best intermediate-fusion baseline. Across five detectors from three architectures, CoMET gains 2.2 to 3.5 AP points over online poses and 0.3 to 0.6 over estimating the offset from the current message alone. We find that the benchmark uses a zero offset as a placeholder in 28% of validation messages and that its offset search reaches the boundary in 45% of the rest. These offsets affect the labels for objects seen only from the roadside. Under two alternative label references, CoMET improves on uncorrected online poses in every evaluated population. Retraining intermediate fusion with CoMET poses and corrected labels improves detection on all three architectures tested, with statistically significant gains on two.
Read the full abstract
Computer Vision
Finite-sample Calibration Guarantees for Binary Segmentation
Machine learning based segmentation models are frequently miscalibrated, producing for each pixel a label prediction whose associated confidence does not accurately reflect the true underlying probability. This limitation is particularly critical in high-stakes applications such as medical image segmentation. Although several calibration methods have been proposed, only a small subset of them provides theoretical guarantees on the calibration error, i.e., they do not typically ensure that, with high probability, the model exhibits a low calibration error. In this work, we develop a post-hoc calibration method for which formal calibration guarantees can be established. To achieve this, we derive from standard concentration inequalities an improved theoretical bound that explicitly accounts for the spatial dependence inherent in binary segmentation outputs. We empirically demonstrate that the proposed calibrator satisfies the derived guarantee and outperforms existing post-hoc calibration approaches.
Read the full abstract
Quantum Physics
Interaction-enabled charging and thermal crossover of the spin-work deficit in a Hubbard double quantum dot
We study how conservation of total spin affects the extractable work of an interacting Hubbard double quantum dot charged in the presence of unresolved quasistatic detuning fluctuations. The full fermionic Fock space is used to model preparation through electronic reservoirs, whereas the additional work deficit associated with spin conservation arises entirely from the singlet and triplet states at half filling. At zero detuning, modulation of the symmetric tunnel barrier couples two singlets of even parity. The corresponding transition matrix element vanishes when U = V . An exact formula for the additional work deficit follows from passive spectral rearrangement under two extraction constraints: conservation of particle number alone and conservation of both particle number and total spin. At low temperature, an expansion in the detuning response gives a leading estimate of the noise amplitude at which this deficit first appears. Increasing the preparation temperature changes the triplet population and produces a crossover at s2 = q, where the additional spin constraint becomes nonbinding for the averaged state, even though both ergotropies may continue to decrease. We obtain a semi-analytic expression for this crossover by separating the thermal Gibbs weights from the singlet dynamics under a fixed charging pulse. Averaging over symmetric detuning disorder also restores parity in the density matrix and allows singlet eigenvalues of different parity to exchange order, changing the branch that determines the thermal boundary. The work values are optimal over specified classes of unitary operations and do not establish the yield of a finite discharge protocol.
Read the full abstract
Image and Video Processing
Semantic Extraction Framework for Goal-Oriented V2X Communication Supporting Safe Intersection Passage
This paper proposes a transmitter-side semantic extraction framework for goal-oriented Vehicle-to-Everything (V2X) communication supporting safe intersection passage. Instead of transmitting raw sensory data, roadside infrastructure extracts task-relevant semantic information from video and traffic-signal sequences and organizes them into compact semantic packets for low-latency driving-decision assistance. The intersection-passing problem is formulated as a set of traffic tasks under Japanese traffic regulations, and the proposed pipeline generates perception and intent semantics through object detection, object tracking, region-of-interest (ROI)-based semantic extraction, and recommended-action generation. The framework is implemented and evaluated on a self-constructed, CARLA-based single-vehicle dataset collected under clear weather conditions. Experimental results show high recommended-action accuracy within the tested task set, while the mean processing latency remains below the frame interval of the 15~fps input video.
Read the full abstract
Computer Vision
PathXAI: Evidence-guided structured pathology reporting by transferable tissue segmentation
Pathology reporting translates tissue observations across regions and slides into diagnoses and supporting morphological descriptions, yet explicit tissue-composition measurements are rarely used to guide H&E-based report generation. We developed PathXAI, which pairs transferable tissue segmentation with a pretrained vision-language model. Because supplying tissue information directly with H&E images gave limited benefit, PathXAI-Report uses segmentation overlays and tissue statistics in separate steps, selects between candidate diagnoses by fixed rules, acquires further images when evidence is insufficient and records every step for review, without updating model weights. Its segmentation component exceeded baselines in mean Dice on internal and external tests. In automated evaluation on 1,200 public and 151 private cases, PathXAI achieved higher agreement with reference diagnoses and documented morphology than H&E-based comparators. On the public cohort, it exceeded Qwen+H&E by 4.4 percentage points in cancer family match rate and 5.7 percentage points in finding F1.
Read the full abstract
Fluid Dynamics
Domain-Adapted Deep Optical Flow for Multiphase Velocimetry in Unsteady Cloud Cavitation
Time-resolved velocimetry inside cloud cavitation is difficult because tracers do not penetrate vapor structures and deforming interfaces distort the particle images that PIV relies on. We present a domain-adapted deep optical flow framework for two-phase velocimetry, a RAFT model fine-tuned on synthetic image pairs rendered from multiphase OpenFOAM simulations (RAFT-CloudCav). It is compared with a pre-trained RAFT model, Farnebäck and OpenPIV on 130,000 frames/s recordings of a Venturi at 30 and 48 L/min. As no experimental ground truth exists, validation uses the integrated flow rate against a turbine meter, the liquid-phase velocity divergence and the station-to-station spread of the flow rate. The four methods agree within 2% in the liquid core and integrate to 41.6-43.2 L/min against a metered 48.0 $\pm$ 0.5 L/min, a shortfall common to all four that lies in the measurement and is partly traced to the coarse imaging configuration. RAFT-CloudCav gives the most consistent flow rate between stations (1.8% against 4.2-4.5%). The learned estimators reduce the liquid-phase divergence three- to seventeenfold relative to the classical ones, with the pre-trained model lowest, and OpenPIV under-reads the cavitating near-wall shear layer by up to 38%. Every method resolves the re-entrant jet in the instantaneous field. In a non-cavitating control the boundary layer is captured better as the imaging is refined, from three grid points across a 0.66 mm layer at 49 $\mu$m/pixel to twelve at 13 $\mu$m/pixel. The mean field is recovered without tracer particles to within 13%. Uncertainties are estimated a posteriori from the signed image residual and propagated conservatively.
Read the full abstract
Algebraic Topology
Infinity operad pairs, higher distributivity and spectral Tambara functors
We introduce an $\infty$-categorical model of operad pairs and their associated bimonoidal categories. We study the related concepts such as cocartesian-cartesian operad pairs, algebra pairs and the Day convolution. Using the Day convolution for operad pairs, we can define distributive functors between a symmetric bimonoidal category and a cartesian closed monoidal category. As an application, we show that for a finite group $G$, the operadic distributive functors from $\mathrm{Span}(\mathsf{Fin}_G)$ to any cartesian closed monoidal category coincides with the distributive functors defined by the distributive diagram. Therefore, we show that the connective $G$-equivariant $\mathbb{E}_{\infty}$-ring spectra are equivalently the spectral $G$-Tambara functors.
Read the full abstract
Representation Theory
Annular webs and Guralnick's conjecture
Using annular webs and Kazhdan-Lusztig mu-coefficients, we disprove Guralnick's absolute boundedness conjecture for first cohomology.
Read the full abstract
Cryptography and Security
Authorization Drift in LLM Agents: Evaluating Boundary Compliance in Multi-Step Tool Use
LLM agents can exceed their authorization boundaries while pursuing user goals without an adversary. We formalize this phenomenon as authorization drift across data, tool, scope, and time, and introduce the Authorization Drift Benchmark (ADB): 648 samples across seven domains, each pairing a sufficient compliant path with an attractive unauthorized shortcut. To ensure executable tasks, tempting shortcuts, and feasible compliant paths, Staged Execution-Reflective Synthesis (SERS) grounds benchmark validation in real agent rollouts rather than LLM self-critique. Across 13 LLMs and three frameworks (25,207 trajectories), greater capability accompanies greater drift, with frontier-capable models exhibiting drift rates of 0.43 to 0.61. Yet 86.18% of tasks admit an observed clean-success trajectory, showing that drift is avoidable. Drift is persistent and dominated by early data over-collection; apparent improvements under agentic scaffolds coincide with collapsing clean-completion performance. To understand potential harms to individuals and society, we map drift points to legal provisions across four jurisdictions, connecting overreach to privacy and information-security concerns. Data minimisation is the most frequently implicated provision, and 93.5% of legally grounded drifting runs map to multiple jurisdictions, making the potential consequences of authorization drift more interpretable.
Read the full abstract
Robotics
Safe Circumnavigation on Heterogeneous Orbits: A Range-Only HOCBF-QP Framework for Nonholonomic Multi-Agent Systems
Circumnavigating a stationary target with a team of robots requires each robot to hold its own standoff distance, to stay evenly spaced from its neighbours, and to avoid collisions with the target and with each other at every instant. The task becomes arduous when the robots are assigned different orbital radii, as their sensors or roles may demand. This paper presents a safety-critical control framework for this problem. Each robot commands both its acceleration and its turn-rate, measures only its distance to the target, and exchanges information only with its two ring neighbours. The design has three features. First, a structural property of the model lets one control law serve every robot regardless of its assigned radius. Second, the acceleration drives each radius exactly to its target, while the turn-rate balances the formation with the neighbours. Third, a higher-order control-barrier-function quadratic program enforces target, geofence, and inter-robot safety on both inputs at once; it leaves the nominal controller untouched away from the constraints and stays valid while the unmeasured range rate is still being estimated. Simulations confirm these guarantees. In a comparison, the closest existing multi-orbit design, even with full state information, violates both the target standoff and the inter-robot margin, whereas the proposed controller does not.
Read the full abstract
Combinatorics
The Sparse-Complement Conjecture for Two Distinct Eigenvalues
For a graph $G$, let $q(G)$ be the minimum number of distinct eigenvalues of a real symmetric matrix whose off-diagonal zero–nonzero pattern is $G$. We prove that $q(G)=2$ whenever the complement $H=\overline G$ of a graph of order $n\ge3$ has at most $n-3$ edges, resolving the sparse-complement conjecture of Fallat and Mojallal. If $H$ contains a cycle, the proof constructs a nontrivial orthogonal projector with the strong spectral property and rank at most $|E(F)|-|V(F)|+4$, where $F$ is the union of the cyclic components of $H$. The construction combines faithful stress-free Gram representations, a one-coordinate lift for each excess core edge, exact orthogonal-frame completion, and a Stiefel-manifold pendant-extension argument.
Read the full abstract
Statistics Theory
Byzantine-Robust Distributed Sparse Softmax Regression
We study Byzantine-robust distributed sparse softmax regression in high-dimensional settings. We propose a communication-efficient distributed algorithm that employs robust aggregation methods—coordinate-wise trimmed mean and coordinate-wise median—to defend against Byzantine attacks. Under sub-Gaussian design assumptions with sparsity constraints, we establish that the distributed estimator achieves the convergence rate of $O_{P}(K^{\frac{5}{2}}\sqrt{s\log(Kd)/N})$ when the Byzantine fraction $\alpha \asymp 1/\sqrt{m}$, matching the centralized oracle rate. Numerical experiments on synthetic data demonstrate the superior performance of the proposed method.
Read the full abstract
Distributed Computing
DARIC: A Position-Independent KV Cache Reuse System with Distribution-Aware Recomputation
The context lengths of large language models (LLMs) are rapidly increasing in real-world applications such as agentic systems and retrieval-augmented generation (RAG). To accelerate LLM inference in these scenarios, reusing position-independent key-value (KV) caches across requests has emerged as a promising approach. This technique typically relies on recomputing a selected subset of tokens to restore cross-chunk interactions. However, existing solutions suffer from critical limitations in both algorithmic efficiency and system deployment. Algorithmically, they overlook the distribution of KV cache deviations, leading to either high token-selection overhead or degraded generation quality. At the system level, the reliance on custom LLM execution flows for recomputation complicates deployment and prevents seamless integration with highly optimized inference engines. In this paper, we propose DARIC (Distribution-Aware Recomputation for Independent KV Cache Reuse), a novel KV cache reuse method based on system-algorithm co-design. At its core, DARIC introduces a distribution-aware token selection strategy that adaptively schedules token recomputation based on empirical KV deviation patterns, striking an optimal balance between generation quality and computational efficiency. Moreover, DARIC is built on a lightweight system architecture that enables independent KV cache generation and recomputation with minimal integration overhead, fully preserving existing system-level optimizations of standard inference engines. Comprehensive experiments across various datasets and LLMs demonstrate that DARIC consistently outperforms state-of-the-art methods, achieving a 1.1$\times$–4.5$\times$ speedup in time-to-first-token (TTFT) with negligible loss in generation quality. Code is available at https://github.com/Lenovo-ICI/DARIC.
Read the full abstract
Optimization and Control
A Smooth Equality Reformulation of Nonlinear Exponential Cone Optimization
Exponential cone optimization has received considerable attention in the modern literature, but studies involving nonlinear objective and constraint functions are still limited. This work can be seen as one of the first steps toward understanding nonlinear exponential cone optimization (NECO) problems. Here, we ask the question "Can smooth equality constraints replace nonlinear exponential cone constraints?". From a practical point of view, even squared slack reformulations in nonlinear programming introduce difficulties. However, such reformulations can be useful for studying optimality conditions. Motivated by this observation, in this work, we use a smooth surjective parameterization of the exponential cone to reformulate NECO as an equality-constrained nonlinear program (RNLP). We then analyze Karush-Kuhn-Tucker (KKT) and second-order optimality conditions for NECO through RNLP. Although feasibility is represented exactly, the KKT conditions for RNLP do not in general guarantee those for NECO. We characterize this correspondence through multiplier restrictions that are equivalent to a second-order condition involving only slack variables. We also establish an exact curvature identity showing that RNLP reproduces the sigma term of NECO without additional assumptions. Finally, we prove the relationships between second-order necessary and sufficient conditions for the two problems. We identify when RNLP can be used to rule out local minimizers of NECO or establish strict local optimality for NECO without explicitly evaluating the sigma term.
Read the full abstract
Computation and Language
PhoPunct: A Large-Scale Benchmark of Transformer Architectures for Vietnamese Punctuation Restoration
Vietnamese punctuation restoration is often evaluated on relatively small benchmarks, limiting conclusions about how architectural choices scale with data volume. We present a controlled benchmark on News and Novels corpora containing 22.1 million labeled tokens (19.9M News and 2.2M Novels). We evaluate 17 architectures formed from four pretrained Transformer backbones and optional bidirectional long short-term memory (Bi-LSTM) and conditional random field (CRF) layers, together with a PhoBERT-large-Bi-LSTM-CRF model, PhoPunct. Every model is trained for eight epochs with the same learning rate and without architecture-specific tuning. Under this fixed budget, adding Bi-LSTM and CRF layers does not reliably improve macro F1: in seven of eight matched domain-backbone comparisons, a plain Transformer classifier matches or outperforms its Bi-LSTM-CRF counterpart. PhoPunct obtains the best News result (61.35), whereas XLM-R with a CRF obtains the best Novels result (50.18), exceeding PhoPunct (48.97). These results show that the value of recurrent and structured-prediction heads depends on their optimization regime and motivate explicit separation between architectural improvements and improvements due to architecture-specific tuning. Our source code is available at: https://github.com/aisia-research-lab/PhoPunct
Read the full abstract
High Energy Astrophysical Phenomena
Probing Dark Matter via the Migdal Effect in LUX-ZEPLIN Data
Sub-GeV dark matter (DM) particles deposit little energy in conventional direct-detection experiments and are therefore difficult to probe, but the Migdal effect provides an electronic signal from atomic ionization accompanying the recoil. Using the new results of the LUX-ZEPLIN experiment with a 4.2 tonne-year exposure, we evaluate the sensitivity of the Migdal effect to the spin-independent DM–nucleon scattering cross section. For masses between $350~\mathrm{MeV}$ and $10~\mathrm{GeV}$, the projected 90% upper limits range from $6.5\times10^{-37}$ to $3.1\times10^{-43}~\mathrm{cm^2}$. These limits are then used to constrain the minimal scalar- and vector-mediator DM models. For the light-mediator DM model, the constraint on the coupling reaches $\alpha\simeq 10^{-12}$, which is stronger than or comparable to the constraints from kaon decays, stellar cooling, and SN1987A.
Read the full abstract
Quantum Physics
Parametric $\mathrm{CZ}$ Gates in a Tunable Tripod Architecture with Fixed-Frequency Superconducting Qubits
Efficient implementation of the operations required for quantum error correction in superconducting quantum computers requires high gate fidelity, adequate qubit connectivity, and selective control of interactions. Sharing a single coupler among multiple qubits can improve connectivity, but also requires accounting for its effects on qubits outside the target pair. In this work, we numerically investigate the feasibility of implementing parametric-resonance $\mathrm{CZ}$ gates in a tunable tripod structure consisting of three fixed-frequency transmon qubits connected to a single tunable coupler. We modulate the external flux applied to the coupler and optimize the pulse parameters for each qubit pair to construct $\mathrm{CZ}_{01}$, $\mathrm{CZ}_{02}$, and $\mathrm{CZ}_{12}$ gates. We calculate the time evolution of a Hamiltonian that includes three energy levels for each circuit element and evaluate gate fidelity and leakage after virtual Z correction. The fidelities obtained for the target-pair computational blocks, whose input and output basis states have both the coupler and spectator qubit in their ground states, are 0.9943, 0.9834, and 0.9907, respectively. We also find that population preservation at the end of the gate depends on the initial state of the spectator qubit. These results demonstrate the possibility of constructing $\mathrm{CZ}$ operations for each qubit pair through flux modulation of a shared coupler while maintaining fixed data-qubit frequencies. They also highlight the need to evaluate performance and optimize control over the full computational space, including the spectator qubit.
Read the full abstract
Optics
Optical microscopy for pathological imaging: past, present, and future
Histopathology, the microscopic examination of stained tissue sections, remains the gold standard for pathology imaging, but its bright-field imaging is largely morphological. Advanced optical microscopy techniques move beyond this by exploiting diverse light-matter interactions, and are increasingly applied to pathological imaging. By capturing signals such as refractive index, polarisation, harmonic generation, Brillouin shift, and molecular vibrations, these microscopy techniques can provide quantitative or semi-quantitative information on tissue structure, mechanics, and chemical or molecular composition. In this review, we summarise the physical principles, key quantitative parameters, and representative pathological applications of seven modalities–optical diffraction tomography, polarisation microscopy, second-harmonic generation microscopy, third-harmonic generation microscopy, Brillouin microscopy, Raman microscopy, and fluorescence microscopy. We also discuss progress in artificial intelligence that supports reconstruction, analysis, and interpretation across these modalities, together with emerging trends in multimodal integration. Overall, this review shows how different optical contrasts can complement routine histopathology, while outlining the challenges and opportunities that will shape their clinical translation.
Read the full abstract
Materials Science
Spontaneous Formation of Lateral Defect-Density Gradient in CVD-Grown MoS2 Monolayers
Defect engineering in two-dimensional materials is a powerful strategy for tuning electronic, optical, and catalytic properties. While pristine molybdenum disulfide (MoS2) has been thoroughly studied, defective MoS2 remains poorly understood due to the complexity and inconsistency of controlled defect creation. Here, we report a defect-density gradient that forms spontaneously in MoS2 flakes during chemical vapor deposition (CVD) growth. Photoluminescence and Raman spectroscopy mapping reveal the lateral arrangement of this gradient across a single flake. Deconvolution of the excitonic photoluminescence emission identifies electron doping as the key factor governing the spatial variation in optical response. This doping likely originates from sulfur vacancies formed during CVD growth, as further supported by the observed Raman mode shifts. We propose that this gradient arises from precursor depletion near the nucleation center during growth, resulting in a gradual variation in defect density. This lateral defect structure provides a convenient platform for systematically characterizing defective MoS2 across a continuous range of defect densities within a single sample. The sulfur-vacancy-driven surface activation identified here can also be important for future catalytic applications.
Read the full abstract
Machine Learning
COMBIO: Automated Discovery of a Compact Composite Biomarker from Routine Laboratory Panels
Machine-learning models based on routine laboratory panels can predict outcomes but can be difficult to interpret and operationalize. We present COMBIO, an automated pipeline that distils such a model into a compact closed-form biomarker. The composite is a product of laboratory measurements raised to integer powers from minus two to two. The sign of each exponent indicates whether higher values of the corresponding measurement increase or decrease the composite. Shapley additive explanation values from the gradient-boosted reference model rank five candidate measurements and guide eligible exponent signs. An exhaustive search over small integer exponents selects the formula on an inner validation set, before evaluation on the reserved test set. We evaluated COMBIO across six tasks: three National Health and Nutrition Examination Survey outcomes, liver-disease classification in the Indian Liver Patient Dataset, and Medical Information Mart for Intensive Care IV in-hospital mortality and sepsis case-finding, with external evaluation in the eICU Collaborative Research Database for the two hospital tasks. Across the six test sets, COMBIO's area under the receiver operating characteristic curve differed by at most 0.03 from logistic regression using the five candidate laboratory measurements. The hospital mortality composite had an area under the curve of 0.698 in the external database. A version restricted to widely available measurements in the source hospital database achieved an area under the curve of 0.786 in a cohort covering 92% of admissions. The multiplicative representation provides compactness rather than additional accuracy, and local validation and recalibration are required before clinical use.
Read the full abstract
High Energy Physics (Phenomenology)
Renormalon-chain contributions to two-point correlators of nonlocal quark currents and light-meson distribution amplitudes
In the framework of QCD sum rules, we study the endpoint slopes of the leading-twist distribution amplitudes in the large-$\beta_0$ approximation for the pion and the longitudinally polarized $\rho$ meson. To this end, we compute the two-point massless correlator of non-local composite vector quark currents with the gluon lines dressed by chains of one-loop fermion bubbles. This correlator depends on the Bjorken fraction $x$ associated with the composite current and, within the large-$\beta_0$ approximation, provides the dominant contribution in each order of perturbation theory. We derive an inequality for the second Gegenbauer moments of these distribution amplitudes that agrees with predictions from lattice QCD.
Read the full abstract
Computer Vision
BayesCrop: Real-Time Probabilistic Multi-Crop Video Reframing
Automatic video reframing is increasingly important for adapting landscape content to heterogeneous display formats. Existing methods typically predict a single deterministic crop, despite the subjective nature of reframing and the presence of multiple plausible regions of interest. We present BayesCrop, a real-time probabilistic framework for multi-crop video reframing. BayesCrop represents crop preferences using a two-component spatially isotropic Gaussian mixture over crop position and scale. Each component defines a representative reframing mode together with its local uncertainty, enabling both dual-view prediction and sampling of additional plausible crop candidates. For efficient video processing, we introduce an anchor-and-propagate strategy that performs neural prediction only on sparse anchor frames and propagates crop trajectories using motion vectors available from the compressed video bitstream. The resulting trajectories can be combined into a single output using a lightweight periodic view-switching strategy. Experiments on LIVE-YT-MC, LIVE-YT-VC++, and a multi-target long-video benchmark demonstrate competitive single-output localization, high multi-crop reference coverage, diverse yet coherent crop hypotheses, and stable perceptual quality. BayesCrop achieves real-time throughput of 112 FPS at the default operating point, while sampling from the learned distributions further improves reference coverage beyond the two representative crop means.
Read the full abstract
Nuclear Theory
Flat "Island of Stability" of Superheavy Nuclei
We present the first systematic microscopic calculations of fission barrier heights for 292 even-even superheavy nuclei using the multidimensionally constrained relativistic mean field model with triaxial and octupole deformations treated simultaneously. The calculated barrier heights remain around 5 MeV in the vicinity of the predicted doubly magic nucleus ${}^{304}120$. This flat landscape contrasts with the conventional picture of mutually enhanced magicity and indicates that the proton shell closure at $Z=120$ and the neutron shell closure at $N=184$ do not reinforce but suppress each other. We introduce an $\alpha$-linked energy difference $\Delta_{\rm pn}$ to identify the mutually suppressed magicity caused by distinct proton and neutron density distributions. These results challenge the expected shell landscape of superheavy nuclei and may have a crucial impact on estimating survival probabilities in the synthesis of superheavy nuclei.
Read the full abstract
Mesoscale and Nanoscale Physics
Disentangling Spin and Orbital Contributions in the Search for Altermagnetic Terahertz Conversion in RuO2
Rutile RuO2 has emerged as a prominent candidate altermagnet, yet the microscopic origin of its reported charge-angular-momentum conversion remains unsettled. Recent theoretical and experimental advances have highlighted a potentially dominant role of orbital transport, raising a fundamental ambiguity between spin- and orbital-mediated conversion in RuO2. Here, we report the first experimental disentanglement of the spin- and orbital-to-charge conversion channels in RuO2 heterostructures using terahertz emission spectroscopy. We identify Fe as a predominantly spin-current injector, whereas Ni and permalloy predominantly inject orbital currents, revealing a substantial orbital contribution to ultrafast terahertz emission. Under controlled spin-current injection, we find no evidence for inverse altermagnetic spin-splitting effect. Instead, the observed anisotropy is consistent with an anisotropic inverse spin Hall response that is strongly conditioned by the near-interface environment. Under orbital-current injection, the anisotropy is markedly weaker, indicating that orbital-to-charge conversion is largely isotropic in RuO2. Our results establish orbital transport and interface-conditioned spin Hall conversion as key to terahertz emission from RuO2, underscoring the need to disentangle spin, orbital, and interfacial contributions before assigning anisotropy to altermagnetism.
Read the full abstract
Statistical Mechanics
$\mathbb{Z}_2$ gauge and exact solution of the $\pm J$ Edwards-Anderson model in a transverse field
Using the Majorana representation of spin-1/2 operators, we study the $\pm J$ Edwards–Anderson model in a transverse field at $T=0$. We show that, in this representation, the quantity describing the coupling between neighboring spins commutes with the Hamiltonian and, unlike in the Kitaev model, is not subject to any Gauss law. The resulting $\mathbb{Z}_2$ gauge freedom allows the disorder in the signs of the exchange integrals to be eliminated by an exact gauge transformation, so that the spectrum of the system coincides with that of the model with uniform couplings. The resulting theory is quadratic in the Majorana fields. Computing the local Green's functions, we find that the Edwards–Anderson order parameter vanishes for $\Delta>J\sqrt{Z}$ and equals $1/4$ for $\Delta<J\sqrt{Z}$, which signals a second-order transition from a paramagnetic phase to a spin-glass ground state that is degenerate over spin configurations and replica symmetric at $T=0$.
Read the full abstract
Computer Vision
Trajectory Forcing: Exploiting Diffusion Trajectories for Autoregressive Long Video Generation
Autoregressive (AR) modeling is arguably the only way to extend a fixed-length video diffusion model to longer or even streaming videos. However, existing AR diffusion methods use only the previously predicted chunk as context for the next, discarding the intermediate predictions that constitute the chunk's generation history. In contrast, conventional AR models such as LLMs condition on the entire history of preceding tokens. To this end, we propose Trajectory Forcing (TraF), an AR diffusion method that exploits the denoising trajectories of previous video chunks as context. TraF operates under two training regimes: (1) with ground-truth videos, where pseudo-trajectories are constructed from Gaussian interpolations and re-predicted by the model, and (2) with a teacher model, where real trajectories are obtained from the student's own denoising rollout. The first regime naturally provides a strong initialization for the second. On the 5-second VBench benchmark, TraF achieves a Total score of 85.15, outperforming Self Forcing (+0.84) and Causal Forcing (+1.11), with Quality and Semantic improving simultaneously. On 30-second generation—6$\times$ the training horizon—TraF achieves the highest overall quality among all compared methods.