Information Retrieval
Think Thrice Before Reranking: Multi-perspective Evidence and Reasoning Integration for Text Reranking
Lijun Liu, Zhengzong Chen, Wenyan Li et al. (+2 more)
Reasoning-based reranking with Large
Language Models (LLMs) has shown promising improvements in text ranking. However, current methods predominantly rely on a single
reasoning trajectory, resulting in rankings that are susceptible to
reasoning errors and inherently constrained in modeling the multifaceted signals underlying document relevance. To resolve this dilemma, we propose MERIT-Rank(Multi-perspective Evidence and
Reasoning Integration for Text Reranking), a framework that models complementary
reasoning trajectories to improve reranking robustness. MERIT-Rank formulates a Multi-Trajectory
Reasoning Space (MTRS) that evaluates query-document relevance from multiple perspectives and introduces a joint reranker that consolidates these
reasoning paths into a unified ranking decision. We further develop Progressive Rank Policy Optimization (PRPO), a progressive training framework that stabilizes
reasoning trajectories while continually improving ranking quality through staged optimization objectives. Experiments on both
reasoning-intensive and traditional
retrieval benchmarks show that MERIT-Rank consistently achieves superior performance over competitive baselines. The 4B model notably outperforms most 7B and even 32B rerankers on BRIGHT.
Read the full abstract ▾
All authors Lijun Liu, Zhengzong Chen, Wenyan Li, Yuanyuan Zhao, Fei Huang
Software Engineering
AdaRepair-Mem: Adaptive Experience Orchestration for Repository-Level Program Repair
Z. C. Luo, J. C. Guo, W. J. He et al. (+10 more)
Recent
memory-augmented repository-level program repair methods reuse historical repair experiences to improve LLM-based issue resolution. However, our analysis reveals three limitations in existing repository-level
memory retrieval. First, episodic
memory is highly imbalanced across repositories, leaving low-resource repositories with little effective support. Second, more
memory does not monotonically lead to higher repair success, suggesting that relevance, quality, and redundancy matter more than raw
memory volume. Third,
memory accumulation is phase-misaligned: repositories may contain many reproduction experiences but few patch or refinement experiences. To address these problems, we propose an adaptive experience
retrieval framework for repository-level program repair. Our framework introduces coverage-aware
retrieval, which falls back to cross-repository or repair-type-based
memories when same-repository
memory is insufficient; quality-aware selection, which ranks
memories by relevance, historical utility, specificity, and redundancy; and stage-aware routing, which separates and retrieves
memories for reproduction, localization, patch generation, patch refinement, and validation. Evaluated on SWE-Bench-Lite and SWE-Bench-Verified, the proposed framework improves repair performance on under-covered repositories, reduces noisy
memory retrieval, and better supports failed-to-fixed patch refinement. Our results show that the key to
memory-augmented repair is not simply accumulating more experiences, but retrieving the right experiences for the right repair context.
Read the full abstract ▾
All authors Z. C. Luo, J. C. Guo, W. J. He, S. Y. Wang, J. C. Yu, F. M. Zhao, Y. Chen, T. Cao, L. Q. Liu, N. Zheng, W. Xu, J. Jiang, Z. M. Zhao
Machine Learning
Local Sparsity Enables Unsupervised LLM Safety Detection
Xin Chen, Gil Kur, Alexander Shevchenko et al. (+1 more)
Deployment-time safety methods for large
language models (LLMs) are predominantly supervised and assume access to unsafe training data. Nevertheless, new attacks and harm categories regularly arise, not captured by models trained in such a supervised fashion. An alternative approach is to view this problem through the lens of anomaly detection, namely, to rely solely on modeling safe data and flagging out-of-distribution inputs. However, LLM activations lie in a high-dimensional space, raising concerns about whether anomaly detection is statistically feasible. We show that, under the linear representation hypothesis (LRH), there may indeed be hope. In the LRH concept space, which is typically recovered via a sparse autoencoder (SAE), nearby points share a small common active support. Using this local sparsity insight, we propose a framework for locally masked SAE-based anomaly detection, supported by theoretical justifications. We validate it on various architectures and
datasets, including both capability-testing
datasets and safety-specific
datasets. Finally, when we allow
algorithms to use 1% out-of-distribution data for calibration, locally sparse methods achieve near-optimal performance, demonstrating their ability to capture meaningful safety information while using only 1-2% of SAE
neurons for computation.
Read the full abstract ▾
All authors Xin Chen, Gil Kur, Alexander Shevchenko, Andreas Krause
Probability
Monte Carlo Approximations of Time-Nonlocal Diffusions in Bounded Domains
Ivan Biočić, Daniel E. Cedeño-Girón, Aleksandar Mijatović et al. (+1 more)
We develop and analyze a Monte Carlo method for sampling killed anomalous
diffusions obtained by time-changing Brownian motion with drift by the inverse of a subordinator. The method targets probabilistic representations of time-nonlocal, including time-fractional, Cauchy–Dirichlet problems on bounded domains. Since inverse subordinators can be sampled exactly in broad classes, while Brownian exit times are generally unavailable in arbitrary domains, we approximate the killed Brownian component by an Euler scheme with discrete boundary detection. We prove a square-root weak error bound with explicit dependence on the Laplace exponent of the subordinator, and derive mean-square and central limit results for the resulting Monte Carlo
estimator. A numerical example in the disk and one in a high-dimensional anisotropic shell illustrate the theoretical rates, computation time, and the mesh-free character of the method.
Read the full abstract ▾
All authors Ivan Biočić, Daniel E. Cedeño-Girón, Aleksandar Mijatović, Bruno Toaldo
Differential Geometry
Retro-synthetic Riemannian geometry in Rocq I : Taylor series of a metric
Gabriella Clemente, Hugo Herbelin and Carlos Simpson
We introduce a synthetic framework for doing local Riemannian geometry that is influenced by a reverse-engineering method for formalizing in Rocq. These new perspectives and techniques lead to a synthetic Taylor formula for a Riemannian metric in normal coordinates, which is the main contribution of this work.
Read the full abstract ▾
Analysis of PDEs
Large-Time Behavior for One-Dimensional Planar Compressible Magnetohydrodynamic Equations on Unbounded Domains
Jing Li and Ziyin Liu
Global existence and the large-time behavior of strong solutions to the one-dimensional planar compressible magnetohydrodynamic system in unbounded domains are investigated.The longitudinal viscosity is assumed to be a positive constant plus any nonnegative power of the density, or to depend on temperature through a power law. The heat conductivity is taken to be proportional to any nonnegative power of temperature, while the transverse viscosity and
magnetic diffusivity are taken to be positive constants. Global existence, uniform-in-time estimates, and convergence to
equilibrium are established without imposing any smallness conditions on the initial data. In the density-dependent case, any fixed nonnegative viscosity exponent is allowed; in the temperature-dependent case, the nonnegative viscosity exponent is required to be sufficiently small in terms of the initial data. An auxiliary
elliptic equation is introduced to control
magnetic-pressure oscillations through a square-root-in-time estimate. Combined with the estimates appropriate to each viscosity law, this yields uniform upper and positive lower bounds for the specific volume and temperature and asymptotic stability of the global strong solution.
Read the full abstract ▾
Solar and Stellar Astrophysics
Investigating the inferred spatiotemporal evolution of magnetic fields during an X1-class flare
Dimitri Arramy, Jaime de la Cruz Rodríguez, Adur Pastor Yabar et al. (+1 more)
Several observational studies have reported sudden temporal changes in chromospheric
magnetic fields inferred during
solar flares using the weak-field approximation (WFA) and nonlocal thermodynamic
equilibrium (NLTE) inversions. We investigate whether these variations reflect the rapid
evolution of the
solar atmosphere or can be influenced by opacity changes caused by
flare heating. We perform spatially coupled and regularized NLTE inversions of high-resolution observations and use the inferred atmospheric models to study the formation of the Ca II 854.2 nm line. We also analyze snapshots from a 3D radiative magnetohydrodynamic
flare simulation to investigate the
evolution of the mapping between column mass and geometrical height. We find that opacity effects alone cannot, in many cases, explain the temporal variations in the reconstructed
magnetic fields, suggesting that they reflect real
solar evolution. However, in approximately 36% of the regions exhibiting rapid
magnetic-field changes, opacity effects are sufficiently strong to bias their detection. Ejected or condensed cold material can also shift the line formation to a different depth, producing an apparent sudden change in the inferred
magnetic field. The simulation further shows that, once the
flare ribbon forms, a given column mass maps to deeper geometrical layers. These results highlight the need for caution when interpreting the temporal
evolution of chromospheric
magnetic fields during
flares, as neither the WFA nor NLTE inversions based on 1D hydrostatic
equilibrium can fully account for these effects.
Read the full abstract ▾
All authors Dimitri Arramy, Jaime de la Cruz Rodríguez, Adur Pastor Yabar, Jorrit Leenaarts
Sound
Multi-Dimensional Prosody Judgment For Live Streaming Speech Synthesis
Zifan Guan, Longyu Lu, Junan Zhang et al. (+3 more)
Evaluating live streaming
speech synthesis (TTS) requires assessing fine-grained, highly expressive prosody such as
emotion, intonation, and energy which traditional MOS predictors fail to capture. While proprietary Large
Language Models (LLMs) like Gemini can evaluate these aspects, they are too costly for massive inference and reinforcement learning feedback. To address this, we first introduce Live-ProsodyJudge (LPJ), a cost-effective pairwise evaluator distilled from Gemini into Qwen3-Omni. However, we identify a critical flaw in standard multi-dimensional evaluation: verdict coupling. The judge tends to lazily align all individual dimension scores with its overall preference, collapsing a rich multi-dimensional rubric into a single preference bit. To resolve this, we further propose Decoupled-Live-ProsodyJudge (D-LPJ). D-LPJ eliminates the overall verdict target to prevent blind following, masks uncertain pair-dimensions during Supervised Fine-Tuning(SFT), and introduces a novel span-local GRPO strategy that applies normalized advantages strictly to their corresponding rationale spans. Evaluated on highly curated human-annotated test sets, 10 sample balanced-order LPJ achieves higher point accuracy than a single Gemini call, while D-LPJ successfully produces independent,decoupled dimension judgments. Furthermore, in a Best-of-8 TTS candidate selection tournament, the LPJ-selected utterance falls within the human top-3 in 85.29% of high-confidence cases, demonstrating its efficacy for fine-grained TTS preference optimization.
Read the full abstract ▾
All authors Zifan Guan, Longyu Lu, Junan Zhang, Zhizheng Wu, Meiguang Jin, Junfeng Ma
Machine Learning
QoS-Aware Federated Learning for Multimodal In-Cabin Interaction in Smart Vehicles
Baran Can Gül, Mert Nak{\i}p, Nasser Jazdi et al. (+1 more)
Modern smart vehicles leverage
multimodal sensors, ranging from high-bandwidth
vision systems to low-rate physiological monitors, to provide personalized in-cabin services. However, integrating high-fidelity
multimodal fusion with collaborative training is often hindered by the heterogeneous and time-varying Quality of Service (QoS) constraints of vehicular networks. Standard
Federated Learning (FL) approaches enforce rigid synchronous rounds that fail to account for these resource asymmetries, leading to safety-critical timing violations and energy exhaustion. In this paper, we propose FedQoS, a novel asynchronous, event-triggered FL framework that decouples local computation from global communication via a two-phase gating mechanism. First, we introduce a resource-aware training gate that initializes local learning only when sensing buffers and energy reserves meet safety thresholds, preventing ML tasks from compromising core vehicle mobility. Second, a QoS-aware transmission policy gates uplink updates based on an efficiency score that balances model novelty against instantaneous latency and energy costs. Locally, clients optimize an objective featuring a staleness-aware proximal term that dynamically adjusts the global anchor strength based on update age. Extensive experiments on
multimodal vehicular
datasets demonstrate that FedQoS achieves competitive personalized accuracy with only marginal performance loss compared to FedAvg, while substantially reducing QoS violations, cutting communication overhead by 76.7%, and lowering latency cost by 26.0%, demonstrating a highly favorable accuracy and efficiency balance for real-world vehicular deployments.
Read the full abstract ▾
All authors Baran Can Gül, Mert Nak{\i}p, Nasser Jazdi, Michael Weyrich
Materials Science
Coherent antiferromagnetic resonance in MnO driven by impulsive terahertz excitation
Yinchuan Lv, Martin J. Cross, Hari Paudyal et al. (+4 more)
Antiferromagnets (AFMs) offer a promising platform for ultrafast information processing owing to their intrinsically fast
spin dynamics and vanishing net magnetization. Realizing this potential, however, requires understanding how terahertz fields excite coherent
magnons and how the resulting
spin motion is transduced into an optical signal. Here we report impulsive terahertz (THz) excitation and time-domain detection of the
antiferromagnetic resonance (AFMR) in single-
crystal manganese(II)
oxide. Time-resolved birefringence measurements reveal long-lived coherent
spin oscillations in the AFM phase. The resonance softens and becomes strongly damped upon warming toward the Néel temperature. Despite this similar excitation behavior, the detected birefringence in MnO is markedly weaker than in NiO. We associate this suppressed optical visibility with the weak
spin–orbit-mediated magneto-optical coupling of orbital-singlet, high-
spin Mn$^{2+}$. These results demonstrate that coherent
magnon excitation and its optical detection are governed by distinct microscopic interactions.
Read the full abstract ▾
All authors Yinchuan Lv, Martin J. Cross, Hari Paudyal, Christopher T. Parzyck, A. H. M. Reid, Durga Paudyal, Matthias C. Hoffmann
Audio and Speech Processing
Alignment-Path Distillation from Non-streaming ASR-LLMs for Streaming Speech Recognition
Yan Jia, Kai Huang, Junjie Chen et al. (+3 more)
In this paper, we propose an
alignment-path
distillation framework for streaming automatic
speech recognition (ASR) with large
language models (LLMs). Interleaved streaming ASR-LLMs use forced
alignments (FA) from
alignment models, such as those trained with connectionist temporal classification (CTC), to construct
speech-text training sequences. However,
alignments obtained from a separate acoustic model may be inconsistent with those learned by LLM-based ASR. This motivates us to transfer
alignment information from a non-streaming ASR-LLM to improve streaming recognition. Specifically, we extract monotonic
alignment paths from a non-streaming teacher's soft text-audio
attention and use them to construct interleaved training sequences. The framework also includes logit and hidden-state
distillation to learn from the teacher's output distributions and internal representations. Experimental results show that, without logit or hidden-state
distillation, training with teacher-derived
alignment paths achieves a 5.2% relative error rate reduction compared with training using forced
alignments. When both models use logit and hidden-state
distillation, teacher-derived
alignments yield a 3.9% relative error rate reduction, with similar mean emission latency but higher flicker. The complete framework achieves a 16.6% relative error rate reduction compared with training using forced
alignments without logit or hidden-state
distillation.
Read the full abstract ▾
All authors Yan Jia, Kai Huang, Junjie Chen, Feng-Long Xie, Xu Tang, Yao Hu
Nuclear Theory
Bayesian calibration of a regional optical potential and uncertainty-quantified predictions for compound nucleus reactions
Samuel Sullivan, Kyle Beyer, Filomena Nunes et al. (+3 more)
Uncertainty-quantified global optical potential parameterizations are useful for making systematic studies across the
nuclear chart and making predictions away from stability. While they generally provide a good description of the reactions on which they were calibrated, predictions can deviate strongly in specific cases and extrapolations incur large uncertainties. We propose a regional optical potential suited for a specific region of the
nuclear landscape. We calibrate the potential along isotopic chains to enable both strong data coverage and improved extrapolative power. We modify the Chapel Hill optical potential and the statistical model used previously by Pruitt et al., to perform a
Bayesian analysis of elastic scattering of
neutrons and
protons from zirconium isotopes with improved data coverage. The parameter distributions of our Chapel Hill Regional Potential (CHiRP) are propagated to compound
nucleus reaction observables including (n, n'), (n, g) and (n, 2n), which have relevance to various
nuclear technology applications. Results are compared to CHUQ, a global calibration of the Chapel Hill potential. The final parameterization of CHiRP differs from CHUQ particularly in the energy and radial dependence of its imaginary components. For elastic scattering, the uncertainties on the observables using CHiRP are smaller than those using CHUQ. CHiRP also provides an improvement over its global counterpart in the agreement with data around 10 MeV, while the regional and global approaches offer similar levels of agreement to elastic data at higher energies. The relative performance of CHiRP and CHUQ when propagating their uncertainty through compound
nucleus reaction channels is also reported. The regional potential approach is a viable alternative to global optical model parameterizations for applications that require precise information within a sub-region of the
nuclear landscape.
Read the full abstract ▾
All authors Samuel Sullivan, Kyle Beyer, Filomena Nunes, Paul Stevenson, James Benstead, Lee Morgan
Algebraic Geometry
On Logarithmic Poisson and De Rham Cohomology Groups of a Class of Inhomogeneous Divisors
Kamtila Kari, Dongho Joseph and Diekouam Fotso Luc Emery
We study logarithmic Poisson and de Rham
cohomologies associated with a class of inhomogeneous free divisors in the affine plane. These divisors arise as inhomogeneous deformations of reduced normal crossing divisors and admit explicit bases for their modules of logarithmic vector fields in the sense of Saito. Using these bases, we construct the corresponding logarithmic Poisson structures and describe explicitly the induced Koszul bracket on logarithmic differential 1-forms within the framework of Lie-Rinehart
algebras. We then determine the logarithmic Poisson cochain complex and compute its
cohomology for the class under consideration. Furthermore, by means of the logarithmic Spencer complex, we identify the corresponding logarithmic de Rham complex and compute its
cohomology. These computations provide explicit cohomological invariants for the class of inhomogeneous divisors considered and show how logarithmic Poisson and de Rham theories extend the corresponding constructions for reduced normal crossing divisors. In addition, we establish an explicit comparison between the logarithmic cohomological theories: they are naturally isomorphic in any degree $k \neq 1$, whereas in degree 1 the logarithmic de Rham
cohomology group is a split one-dimensional extension of the logarithmic Poisson
cohomology group.
Read the full abstract ▾
Quantum Algebra
Arnab Bhattacharjee
We introduce the
Hopf image of a right coaction of a
Hopf algebra on an
algebra as the smallest
Hopf subalgebra through which the coaction factors. We establish its basic properties and show that the induced coaction on the
Hopf image is inner-faithful. For coactions on Nichols
algebras arising from Yetter–Drinfeld modules, we prove that the
Hopf image is the smallest
Hopf subalgebra detected by the underlying coaction on the Yetter–Drinfeld module. We give explicit examples of proper non-trivial
Hopf images arising from bosonizations.
Read the full abstract ▾
Instrumentation and Detectors
Commissioning and Performance of the Time-of-Flight Detector for the T2K Neutrino Oscillation Experiment
C. Alt, L. Amziane, N. Baudis et al. (+23 more)
The T2K ND280 Upgrade aims to reduce systematic uncertainties in measurements of
neutrino oscillation parameters and improve sensitivity to the charge-parity (CP)-violating phase, $\delta_{\mathrm{CP}}$. A key component is the Time-of-Flight (ToF) detector, comprising six panels with 118 EJ-200 plastic-scintillator bars surrounding the Super Fine-Grained Detector (SuperFGD) and two High-Angle Time Projection Chambers (HA-TPCs). Each bar is read out at both ends by silicon photomultiplier arrays and digitised using SAMPIC waveform electronics. The ToF provides precise timing and particle-direction information for particle identification and rejection of backgrounds entering the tracker from outside. This article presents the detector design, construction, signal reconstruction, integration, commissioning, and performance. Dedicated single-bar measurements achieve a time resolution of approximately 130 ps and a longitudinal position resolution of 2.6 cm. After installation in ND280,
cosmic-ray calibration yields an in situ single-bar time resolution of $169 \pm 1$ ps for Top-Bottom crossing events. Beam data clearly resolve the eight-bunch T2K spill structure, confirming synchronisation with the ND280 trigger and data-acquisition systems. The ToF has been successfully commissioned and operates stably within the upgraded ND280 detector.
Read the full abstract ▾
All authors C. Alt, L. Amziane, N. Baudis, A. Blanchet, S. Bordoni, T. H. Bui, F. Cadoux, P. Collard, M. El Baz, Y. Favre, L. Giannessi, G. Ha, C. Jesús-Valls, V. S. Kasturi, A. Klustová, A. Korzenev, T. A. Le, T. Lux, A. D. Nguyen, D. T. Nguyen, H. Nguyen, S. Samani, F. Sánchez, M. Ta, T. Thaiduc, E. Villa
Robotics
How Far Can GPT-6-Astra Go? Evaluating Capabilities in Zero-Shot Vision-and-Language Navigation
Guangzhao Dai, Qi Wu and Bin Zhu
We study GPT-6-Astra in a zero-shot
Vision-and-
Language Navigation in Continuous Environments (VLN-CE) system, where it interprets instructions, assesses its surroundings, and proposes actions. The system uses a common observation–decision–execution workflow with direct model API calls, without a packaged
agent harness or
navigation-specific fine-tuning. In this workflow, each request receives selected observations, execution feedback, and retained progress records. Evaluation covers the complete system, including context management and action control. We evaluate the system on 50 of the 100 R2R-CE val-unseen episodes used by Open-Nav. It achieves a success rate of 52.0%, an SPL of 48.9%, and an nDTW of 70.8%. Our analysis highlights three findings. First, recorded responses link landmarks and earlier actions to instructions using observations and supplied history. Second, reviews include requests for additional views and revisions of uncertain judgments. Third, the results suggest a gap between task understanding and autonomous completion: an unfinished crossing is recognized while rotation continues. At termination, 36.0% of episodes succeed with a workflow-accepted STOP, while another 16.0% meet the distance criterion at the step limit. These results highlight a central challenge: translating correct local judgments into sustained progress and appropriate stopping.
Read the full abstract ▾
Systems and Control
The Small-Talk Effect in Practical Synchronization of Heterogeneous Oscillating Dynamics
Finn Voland, Vincent Schmidtke, Zonglin Liu et al. (+1 more)
This paper investigates multi-time scale commu- nication schemes in synchronization of heterogeneous Lienard oscillator systems. Existing results rely on global exchange of information for all networked systems at discrete time instants, providing limited flexibility in adjusting the trade-off between communication rate and synchronization performance. A scheme to investigate how global updates are complemented by more frequent local interactions within a sub-network is thus proposed in this paper. For a prescribed bound on the synchronization error, conditions are derived under which the global and local update rates can be selected as design parameters. A numerical example illustrates the beneficial effect of the local interactions.
Read the full abstract ▾
All authors Finn Voland, Vincent Schmidtke, Zonglin Liu, Olaf Stursberg
Robotics
Sarvesh Prajapati, Ananya Trivedi, Lorena Maria Genua et al. (+3 more)
Collecting real-world
navigation data for mobile
robots typically requires platform-specific teleoperation, making large-scale collection expensive and difficult to scale. We introduce Universal
Navigation Interface (UNI), a
robot-free data collection paradigm that uses a four-wheeled rollator walker (rollator) and smartphone to collect physically constrained human demonstrations. Because the rollator cannot climb stairs, negotiate uncut curbs, or pass through narrow gaps, demonstrations are naturally biased toward wheeled-feasible routes. Using UNI, we collect 37.2 km of real-world
navigation data and recover metric trajectories that directly supervise goal-conditioned
navigation models. Fine-tuning visual-
navigation models on UNI reduces trajectory prediction error by 17.4-24.8% on held-out UNI demonstrations. Evaluation on other
navigation datasets shows benefits that vary by
dataset and metric. We further demonstrate closed-loop transfer to a powered wheelchair in curb, staircase, and curb-cut scenarios. These results support low-cost physical proxies as a practical source of
navigation supervision collected without the target
robot.
Read the full abstract ▾
All authors Sarvesh Prajapati, Ananya Trivedi, Lorena Maria Genua, Drake Moore, Bruce Maxwell, Taskin Padir
Statistical Methodology
Identification and Estimation of Causal Estimands with Missing Not at Random Data
Faria Rauf Ria, Tarikul Islam and Mahbub A.H.M. Latif
Missing not at random (MNAR) data pose significant challenges for causal inference, particularly when both confounders and the outcome are partially observed. Without additional assumptions beyond those required for causal inference, causal estimands are generally not identifiable under MNAR mechanisms. This paper first develops identification results for the causal estimand, the average treatment effect, under several plausible MNAR mechanisms using completeness conditions, and proposes an estimation approach based on the Expectation-Maximization (EM)
algorithm. We further extend this identification and estimation framework to mediation analysis, enabling the estimation of natural direct and indirect effects under MNAR mechanisms. Through extensive simulation studies, we compare the proposed method with two widely used approaches for handling missing data, complete-case analysis and multiple imputation. The results show that the proposed method yields substantially lower bias under the considered MNAR mechanisms. Finally, we apply the proposed approach to NHANES data to estimate the causal effect of education on depression, with health condition as the mediator.
Read the full abstract ▾
High Energy Physics (Theory)
Supersymmetric boundary algebras in AdS$_3$ gravity: constraints, Dirac brackets and non-locality
Nabamita Banerjee, Vedant Bhutra, Suvankar Dutta et al. (+1 more)
We study supersymmetric boundary dynamics in AdS$_3$
gravity by treating boundary conditions as constrained reductions of a generic Chern-Simons boundary phase space. We focus on how the constraint structure controls the locality and physical field content of the reduced asymptotic
symmetry algebra. For supersymmetric Kac–Moody boundary conditions, the surviving bosonic currents remain local, whereas the
fermion–
fermion Dirac bracket contains the inverse of a field-dependent first-order differential operator and is therefore intrinsically non-local. Imposing an additional fermionic constraint reveals a further structure: after the bosonic second-class reduction, it becomes first class and generates a fermionic gauge redundancy. Generically the remaining fermionic field can be gauged away, while special backgrounds with periodic zero modes support non-trivial physical fermionic sectors. We further employ the Batalin–Fradkin–Tyutin formalism to eliminate the non-locality of the reduced
symmetry algebra by converting the second-class constraints into an equivalent local first-class system on an enlarged phase space.
Read the full abstract ▾
All authors Nabamita Banerjee, Vedant Bhutra, Suvankar Dutta, Soumava Kundu
Astrophysics of Galaxies
Electron irradiation and temperature-programmed desorption of layered ethanolamine-water-methanol ices
Tara L. Stoib, Alejandro Guerrero-Caicedo, Duncan V. Mifsud et al. (+12 more)
Ethanolamine (EtA) is a recently detected interstellar complex organic molecule and a potential precursor to more functionalised N-bearing species. We investigate the electron-driven chemistry of layered EtA–H$_2$O–CH$_3$OH ices deposited at 20~K, combining 2~keV electron irradiation with subsequent temperature-programmed desorption (TPD) while monitoring the solid phase by mid-infrared spectroscopy. These controlled irradiation conditions represent a simplified laboratory analogue of the structural and energetic conditions within interstellar grain mantles. Volatile products released during warming were followed by quadrupole mass spectrometry, and the final residue was analysed by ex situ electrospray ionisation mass spectrometry (ESI-MS). Irradiation produces abundant radiolysis products typical of processed methanol- and water-rich ices, including CO, CO$_2$, H$_2$CO, H$_2$O$_2$, NH$_3$, and OCN$^{-}$. Bands consistent with formic acid and polyoxymethylene-like material are also observed, indicating that carbonyl chemistry and formaldehyde-driven oligomerisation can occur alongside fragmentation in the layered system. Comparison with a choline chloride–H$_2$O reference reveals the emergence of features compatible with methylated EtA derivatives and/or choline-like species, while ESI-MS shows higher-$m/z$ mass families in the residue that are not reproduced in the substrate/background blank. However, spectral congestion and
matrix effects preclude an unambiguous assignment of choline. Our results show that energetic processing of EtA-containing ices can generate a diverse inventory of O- and N-bearing products, providing laboratory constraints on radiation chemistry in cold, irradiated grain mantles under idealised conditions.
Read the full abstract ▾
All authors Tara L. Stoib, Alejandro Guerrero-Caicedo, Duncan V. Mifsud, Péter Herczku, Gergő Lakatos, Zuzana Kaňuchová, Zoltán Juhász, Béla Sulik, Nigel J. Mason, Sergio Ioppolo, Paola Caselli, Barbara Michela Giuliano, Karl E. Duderstadt, Felipe Fantuzzi, Heidy M. Quitián-Lara
Artificial Intelligence
Perception, Layout, and Validation: Calibrated Confidence for Reliable Straight-Through Processing of Financial Documents
Yichao Jin, Yushuo Wang, Yuxuan Han et al. (+7 more)
Straight-through processing (STP) on extracted key-value fields from financial documents without human review requires a calibrated probability together with a bounded guarantee on the residual error of the auto-approved tier. The emergence of modern
Vision Language Models (VLMs) provides an out-of-the-box capability for extracting the key-values, but their verbalized confidence signals are unreliable and weakly track field correctness. This paper introduces a decomposed confidence layer along three interpretable channels, including perception, layout, and validation. Together with a final conformal risk control, the score can be used for reliable STP of financial documents. The method is validated on three public
datasets covering real invoices, synthetic invoices, and ad-buy forms, using two different VLM families (Qwen3.6-27B and Gemini-3.1-Flash-Lite). Our decomposed score consistently improves the separation of correct from incorrect extractions, substantially raising the AUROC from 0.54-0.74 for VLM verbalized signals to 0.90-0.99 with contributions from all three designed channels. Crucially for industrial deployment, this enables usable STP. The native VLM confidence signals could clear only 0.1%-7.0% of fields under risk control at a target error of <10%. In contrast, the proposed method auto-approves 49-72% of fields while holding the empirical error of the accepted tier at or below the target.
Read the full abstract ▾
All authors Yichao Jin, Yushuo Wang, Yuxuan Han, Kwan Ching Yee Sonia, Weiyang Song, Chiu Jin-Chun Kent, Wong Chong Hwee, Wong Tiong Kiat, Kenneth Zhu Ke, Jingyuan Zhao
Analysis of PDEs
Stability for the 2D Micropolar equations near Couette flow via Green's function method
Jie Qi and Lei Yu
In this paper, we use the Green's function method to study the stability of solutions to micropolar equations in the whole space. An essential difficulty in estimating the Green's function lies in treating a coupled system of equations with variable coefficients, which is completely different from previous Green's function analyses for scalar equations. We propose a dominant factor extraction method to overcome this difficulty. Based on sharp Green's function estimates for the micropolar system, we prove that if the initial data (m_0, \omega_0) satisfy \|(m_0, \omega_{0})\|_{L^1\cap L^{\infty}}\leq c_0\mu^{3/4} for some small constant c_0 independent of the viscosity coefficient \mu, then the corresponding solution remains of order O(\mu^{3/4}) under the effect of the Couette flow. Meanwhile, we establish the optimal decay estimates of solutions in the general L^p norm. More importantly, the dominant factor extraction method developed in this work provides a general framework for the stability analysis of complex coupled fluid systems.
Read the full abstract ▾
Classical Analysis and ODEs
$L^p$ Decay Estimates for Circular Means of Fractal Measures in $\mathbb{R}^2$
Zhenbin Cao, Feilong Guo and Junfeng Li
In this paper, we investigate weighted $L^p\rightarrow L^q$ restriction estimates for the
Fourier extension operator associated with planar curves of non-vanishing
curvature. More precisely, we establish estimates of the form \begin{equation} \|Ef\|_{L^q(X)}\leq C_{\epsilon}R^\epsilon\|f\|_p, \end{equation} where $X$ is an $\alpha$-dimensional set. As an application of this result, we derive a new $L^p$ decay estimate for circular means of
Fourier transforms of fractal measures in $\mathbb{R}^2$. We also construct examples that give several upper bounds for the decay exponent when $p\in[1,2]$.
Read the full abstract ▾
Robotics
AnyViewDex: View-Invariant Dexterous Manipulation from RGB Observations
Soham Patil, Om Sanjay Gunjal, Sourabh Bhosale et al. (+3 more)
Visuomotor policies for multi-fingered dexterous manipulation are highly sensitive to camera viewpoint shifts. To achieve view invariance, recent methods increasingly rely on explicit 3D modalities like RGB-D or point clouds, which can introduce hardware dependencies, calibration requirements, and vulnerability to sensor noise during real-world deployment. In this work, we show that view-invariant control can be achieved without explicit test-time 3D sensing by encoding
geometric knowledge into the visual representation during simulation. We present AnyViewDex, an asymmetric training pipeline that combines multi-view contrastive
alignment with privileged 3D
geometric supervision. By regressing absolute 3D object coordinates during simulated training, this auxiliary objective provides a
geometric grounding signal that mitigates the spatial collapse of the globally pooled contrastive
embedding. At deployment, the policy operates zero-shot using only uncalibrated
monocular RGB and proprioception. We validate this approach across both reinforcement learning and student-teacher
distillation. In hardware evaluation on an xArm7 with a 16-DoF LEAP Hand, AnyViewDex reaches 76.7% grasping success across eight unseen objects and six uncalibrated viewpoints (480 trials; 2,400 across all ablation conditions), indicating that geometrically grounded
monocular policies transfer zero-shot without test-time depth. Project Page: https://anyviewdex.github.io/
Read the full abstract ▾
All authors Soham Patil, Om Sanjay Gunjal, Sourabh Bhosale, Arhan Chavare, Ramandeep Singh Hora, Spandan Roy
Computer Vision
AnyviewMeter: Adapting Robotic Reward Models with Camera Geometry and Multi-View Attention
Yuang Tu, Runjia Tan, Yujie Yan et al. (+2 more)
Robotic reward models evaluate task execution from visual observations, but their predictions can change with camera viewpoint and occlusion even when the underlying task state is unchanged. Adapting a pretrained reward model to a local task therefore requires accounting for how that task is observed. We introduce AnyviewMeter, a geometry-conditioned adaptation framework for robotic reward models that represent task progress as a scalar reward signal. It combines low-rank fine-tuning with
token-aligned Plucker rays and synchronous block
attention: ray conditioning incorporates camera geometry into visual features and
attention queries and keys, while block
attention fuses synchronized views inside the pretrained decoder. The framework supports both single-view reward prediction and joint multi-view evaluation through parameter-efficient adaptation of a pretrained Robometer model. On PickCube, single-view adaptation improves progress prediction in every camera group and reduces mean absolute error under a changed field of view by approximately 21% relative to RGB fine-tuning. Across simulated manipulation tasks, joint multi-view prediction reduces progress error by 41-69% compared with averaging single-view RGB predictions and improves temporal ordering in approximately 88% of task-camera groups. On real tasks with fixed and wrist-mounted cameras, mean absolute error decreases by approximately 21% relative to averaged RGB fine-tuning. These results support camera geometry and joint visual evidence as useful components of task-specific robotic reward adaptation.
Read the full abstract ▾
All authors Yuang Tu, Runjia Tan, Yujie Yan, Jinghan Hu, Chen Lv
Image and Video Processing
Effects of Sequence Timing on the Spatio-Temporal Properties of 3D BOLD fMRI: A Formal Framework and Analysis
Samuel Bianchi and Klaas P. Pruessmann
3D sequences provide an alternative to 2D or multiband sequences for BOLD fMRI. The impact of acquisition time differences between slices is well understood for 2D or multiband sequences. For 3D sequences, k-space is partitioned into multiple segments and the final image depends on samples taken over an extended period of time. Any precise timing information is lost during reconstruction. A theoretical description of how sequence timing impacts 3D BOLD fMRI data properties is lacking in the literature. We present a formal framework that models sequence timing effects and draw connections to existing literature. The framework rests on the statistical description of a
spin system that allows incorporation of BOLD signal changes and is completed by a general description of spatial encoding and image reconstruction. Using this formulation, we define three key images: the actual image reconstructed from a segmented 3D acquisition, an optimal reference image unaffected by sequence timing, and an error image representing their difference. Two sets of operators are defined that describe the generation of these images and allow us to analyse the effects of sequence timing independent of the object. All operators take the form of spatio-temporal filters. They mix signal content spatially and based on its temporal waveform. High-frequency BOLD signal content is affected more strongly compared to low-frequency content. Simulations demonstrate that while the total variance of 3D image time series can be reduced, spatio-temporal fidelity is increasingly lost. Furthermore, we hypothesize that the heightened sensitivity of 3D sequences to physiological noise (compared to 2D) is caused by sequence timing damping thermal noise more strongly than physiological fluctuations. The impact of motion and system instabilities on image time series needs to be studied further.
Read the full abstract ▾
Computation and Language
Design of the IBM Granite 5.0 TurboCTC ASR Model
Brian Kingsbury, George Saon, Masayuki Suzuki et al. (+5 more)
We describe the architecture, training methodology and inference speedups of Granite 5.0 Turbo CTC, a 470 million parameter encoder-only model with an excellent speed-accuracy tradeoff. The architecture uses pyramidal temporal subsampling within Conformer blocks using strided depthwise convolutions, block-diagonal (chunk-wise) self-
attention, and conditioning on intermediate predictions from the middle layer. Training highlights are the use of only publicly available data, the novel use of a
Muon optimizer, and balanced data sampling. Inference speedups include replacing 1 x 1 convolutions with linear layers and optimizing the
attention computation in the Conformer blocks. Collectively, these result in a model that is on the speed-accuracy Pareto frontier of the Open ASR leaderboard for English short-form ASR while being twice as fast as the fastest competitor. The model can be used under a permissive license and downloaded from https://huggingface.co/ibm-granite/granite-
speech-5.0-470m-turboctc.
Read the full abstract ▾
All authors Brian Kingsbury, George Saon, Masayuki Suzuki, Hong-Kwang J. Kuo, Takashi Fukuda, Samuel Thomas, Vishal Sunder, Avihu Dekel
Robotics
Safety-Critical Scenarios Emerge from Initial Scenes
Yin Wu, Jiarong Wei, Carl Esselborn et al. (+4 more)
Safety-critical driving scenario generation has largely focused on manipulating the behavior of surrounding
agents while starting from an initial scene from driving data. This assumption can limit the space of discoverable failures, since driving data can provide little opportunity for meaningful interaction. For example, in the Waymo Open Motion
Dataset, 20.44% of recorded slices feature a stationary ego vehicle that never moves, and 30.39% of initial frames contain no nearby traffic participants within 10 meters. We instead study safety-critical scenario generation as an initialization problem: given agnostic black-box driving policies, we learn to generate realistic initial scenes that are more likely to evolve into critical interactions. We propose AdvScene, a conditional latent
diffusion model that is trained in two stages. Starting from pretraining on naturalistic driving data, we post-train the
adversarial-
agent generation branch using reinforcement learning with feedback from closed-loop simulator rollouts. Conditioning on ego driving displacement prevents the ego from remaining static, and RL finetuning induces criticality directly with non-differentiable safety-critical metrics. Experiments on the Waymo
dataset across 12 combinations of ego and traffic policies show that our AdvScene substantially increases the rate of ego-fault collision events and TTC<3s events.
Read the full abstract ▾
All authors Yin Wu, Jiarong Wei, Carl Esselborn, Shubham Phoolari, Ahmed Abouelazm, Daniel Slieter, J. Marius Zöllner
Instrumentation and Methods for Astrophysics
The SHARPEx space-to-space VLBI experiment
Freek Roelofs, Christiaan Brinkerink, Albert Catalan Artigas et al. (+16 more)
Space is the next frontier for ultra-high resolution imaging of astrophysical sources with very long baseline interferometry (VLBI). The limited size of the Earth and atmospheric absorption and
turbulence at high frequencies prevent order-of-magnitude resolution improvements from current black hole imaging with the ground-based Event Horizon
Telescope. A two or three-element submm space-based array like SHARP would image the thin black hole
photon ring and measure the black hole
spin and spacetime. Here, we present the SHARP Experiment (SHARPEx) mission concept. SHARPEx will demonstrate key enabling technologies for space-to-space VLBI, including clock syntonization between two satellites, on-board correlation, and sub-wavelength relative orbit determination. SHARPEx will consist of two CubeSat or SmallSat platforms carrying small (~1 m diameter)
antennas, observing at cm wavelengths. While SHARPEx will be a technical demonstrator, a SHARPEx-type follow-up mission with enhanced intersatellite link properties (range and bandwidth) may produce images of bright AGN with unprecedented resolution and fidelity.
Read the full abstract ▾
All authors Freek Roelofs, Christiaan Brinkerink, Albert Catalan Artigas, Heino Falcke, Jaime Fernández, Christian M. Fromm, Felix Glaser, Marc Klein Wolt, Yuri Kovalev, Volodymyr Kudriashov, María Manzano, Rasmus Maråk, Manuel Martin-Neira, Jacobo Pastor Fernández-Posse López, Montserrat Puertolas Turell, Alberto Rodríguez Pérez-Silva, René Rüddenklau, Efthalia Traianou, Roger Vilaseca Miro