Analysis of PDEs
Optimal Hardy Inequalities for Fractional $p$-Laplacians on the Integers
Florian Fischer and Marius Nietschmann
We obtain optimal Hardy inequalities for fractional powers of the $p$-
Laplacian $\Delta_p^\sigma$ on $\mathbb{Z}$, $p\in (1,\infty)$, $\sigma \in (0, 1/p)$. Asymptotically, the optimal weights behave like $|x|^{-p\sigma}$ as $|x|\to \infty$. We show a similar result for the Riesz fractional $p$-
Laplacian. In an appendix, we show that for general $p$-Laplace operators on locally summable graphs, $p$-null-criticality implies $p$-optimality at infinity.
Read the full abstract ▾
High Energy Physics (Experiment)
First Measurement of the Superscaling Variable as a Function of Visible Kinematics in Charged-Current Quasi-Elastic Neutrino–Nucleus Scattering
The MINERvA Collaboration: L. Giannessi, S. Akhter, Z. Ahmad Dar et al. (+32 more)
We report the first measurement of the superscaling variable $\psi^{\prime}_{\rm vis}$ in
neutrino–
nucleus scattering, using charged-current quasielastic-like $\nu_\mu$-Carbon interactions collected by the MINERvA experiment in the medium-energy NuMI beam. Originally developed to characterize phenomena of scaling with respect to atomic number and momentum transfer in quasielastic electron-
nucleus scattering, $\psi^{\prime}$ is reconstructed here from visible final-state kinematics ($\psi^{\prime}_{\rm vis}$). $\psi^{\prime}_{\rm vis}$ is used here as an externally calibrated variable whose peak position is sensitive to the invisible
nuclear missing energy. We measure $\psi^{\prime}_{\rm vis}$ as functions of the hadronic
recoil $\Sigma T_p$ and
muon transverse momentum $p_{\text{T}\mu}$, comparing background-subtracted data to five
neutrino–
nucleus interaction models via a forward-folding technique. The predicted $\psi^{\prime}_{\rm vis}$ diverges most strongly among model predictions at low $\Sigma T_p$ and $p_{\text{T}\mu}$, where agreement with data is also poorest, exposing which models fail to capture the phase-space dependence of the missing energy and therefore the
neutrino energy as extrapolated from visible kinematics. Among tested models, we find that the Relativistic Mean Field approach predicts more accurately the $\psi^{\prime}_{\rm vis}$ peak position.
Read the full abstract ▾
All authors The MINERvA Collaboration: L. Giannessi, S. Akhter, Z. Ahmad Dar, N.S. Alex, M. Sajjad Athar, J. Felix, L. Fields, R. Gran, E.Granados, D.A. Harris, A. Klustová, M. Kordosky, D. Last, and A. Lozano, S. Manly, W.A. Mann, K.S. McFarland, M. Mehmood, O. Moreno, J.G. Morfín, J.K. Nelson, G.N. Perdue, C. Pernas, M.A. Ramírez, R.D. Ransome, N. Roy, D. Ruterbories, F. Sánchez, H. Schellman, C.J. Solano Salinas, D.S. Correia, V.S. Syrotenko, N.H. Vaughan, A.V. Waldron, L. Zazueta
Robotics
Taming VLAs under Robot Execution Errors: Self-Compensation and Stress Testing
Sohyun Lee, Yoonjae Baek, Jaesang Won et al. (+5 more)
Vision-
language-action (VLA) policies often fail when a
robot's executed motion deviates from their commanded action. Such execution errors arise from the
robot's mechanics and operating conditions, such as wear and payload changes. We propose self-compensating VLA, a deployment-time adaptation method that enables a VLA policy to pre-compensate for the
robot's execution errors when generating commands. Without task rewards or labels, it updates the policy online using the residual between the action commanded by a VLA and the motion executed by the
robot. To stress-test VLA robustness across execution conditions that are impractical to cover with physical
robots alone, we introduce RoboStress, a controlled simulation
benchmark. It combines established joint-level models of friction, backlash, compliance, and
gravity-compensation error into seven deployment scenarios whose execution errors depend on the
robot's state and motion history. On RoboStress, self-compensating VLA achieves higher average task success than both the base policies and methods that build in robustness during training. On two physical
robot arms with different usage histories, it raises the average task success rate by more than 30 percentage points on each arm, and the gains extend to objects not seen in the task demonstrations.
Read the full abstract ▾
All authors Sohyun Lee, Yoonjae Baek, Jaesang Won, Jinnyeong Kim, Kang Hyunwoo, Seung-Hwan Baek, Ivan Laptev, Suha Kwak
Mathematical Physics
Tripartite System Dynamics of Cooperation Ecological Resources and Adaptive Regulation
Linjie Liu, Yike Li and Shijia Hua
To address commons dilemmas under costly social cooperation, this paper develops a tripartite coevolutionary system that integrates social behavior, ecological resource dynamics, and institutional regulation. Institutional intervention is modeled as an adaptive feedback mechanism whose intensity responds to deviations from prescribed cooperation and ecological targets. Through
equilibrium analysis, stability analysis, and phase space characterization, we identify the principal operating regimes and transition mechanisms of the coupled system. The results show that the system admits a sustainable operating
equilibrium in which cooperation, ecological resources, and institutional regulation coexist. We further establish the structural stability of a desirable operating state characterized by full cooperation, peak resource availability, and persistently minimal intervention. In addition to these stable regimes, the system exhibits bistability and limit cycles, indicating that alternative long term outcomes may arise from different initial conditions and parameter configurations. These nonlinear mechanisms generate a persistent crackdown rebound policy trap, in which intensified intervention temporarily improves cooperation and resource conditions but eventually produces recurrent oscillations and prevents system wide consensus. The proposed framework provides a system level explanation of governance failure and offers analytical insights for designing adaptive institutional regulation in coupled human environment systems.
Read the full abstract ▾
Number Theory
Truncated pretentious distances of multiplicative arithmetic functions
Thomas Renard
Let $f\colon\mathbb{N}\to\mathbb{U}$ be a non-pretentious multiplicative function in the sense of Granville and Soundararajan. We study the convergence of the autocorrelation averages of $f$. In particular, we investigate the relationship between the local pretentious behaviour of $f$ on intervals of the form $[x^{\varepsilon},x]$ and its deviation from $1$ on the
primes. As an application, we exhibit a class of non-pretentious multiplicative functions for which Elliott's
conjecture in its original formulation holds, following up on a result by Klurman, Mangerel and Teräväinen.
Read the full abstract ▾
Computer Vision
OFBD: Object-Focused Background Debiasing for Long-Tailed Learning
Shenghan Chen, Yiming Liu, Zhipeng Deng et al. (+6 more)
Balancing performance trade-offs on long-tailed data distributions remains a long-standing challenge in visual recognition. Existing methods mainly improve tail classes through re-balancing, representation learning, or data augmentation, but the underlying cause of tail class degradation is still insufficiently explored. In this paper, we find that standard long-tailed training induces background-biased representation and optimization: tail classes suffer larger background distribution shifts and become increasingly driven by background gradients. This reveals that tail degradation is not merely caused by insufficient samples, but also by the learning of irrelevant background features. To tackle this issue, we propose Object-Focused Background Debiasing (OFBD), a framework that mitigates background bias from both distribution and optimization perspectives. Specifically, Foreground-guided CutMix preserves target-related foregrounds while diversifying complementary backgrounds, and Background-guided Feature Rectification suppresses background-biased features without learnable parameters or additional training. Extensive experiments show that our method improves overall accuracy, achieves significant tail-class gains, and can serve as a plug-in for mainstream long-tailed methods without external data or pretrained recognition models. The code is available at: https://ofbd-neurips2026-longtail-learning.github.io/
Read the full abstract ▾
All authors Shenghan Chen, Yiming Liu, Zhipeng Deng, Haolin Wang, Jiale Zhou, Zhijian Wu, Xiankai Lu, Yafei Ou, Yefeng Zheng
Computer Vision
The Domain Is a Residue: Adapting Self-Supervised Features, Not Generators
Thomas Deixelberger and Markus Steinberger
Clearing fog, rain or snow from footage, or turning renders into photographs, must remove the source domain and keep the scene. Unpaired translators carry it through because their generator sees the source appearance (pixels, a near-invertible latent or a control map) and keeps it. A DINO feature map fixes what is in the scene and carries weather, lighting and rendering style as a residue of 13 to 14% of the feature norm. We propose the Representation Feature Adapter (RFA), a 2.9M-parameter network that moves this residue. We train only the adapter and its discriminators; the encoder and a feature-conditioned decoder, trained once for all conditions, stay frozen. Against CycleGAN-Turbo it is ahead on both metrics on fog and on KID on night, and level within noise on snow, rain and haze. On sim-to-real it leads REGEN and HyPER-GAN on both metrics. Only the RFA removes the rain while keeping the scene. The removal costs scene structure: CycleGAN-Turbo keeps more on every condition but fog. On VAE latents the identical adapter collapses to the identity, and decoders from other groups that never saw it render its output. The RFA has about 160 times fewer trainable parameters than CycleGAN-Turbo and under a fifth of its per-condition training time.
Read the full abstract ▾
Human-Computer Interaction
EntityWeaver: Visual Exploration and Curation of Named-Entity Relationships in Document Collections
Uroš Šmajdek and Ciril Bohak
We present an interactive visualization system for exploring named entities and their relationships across document collections, with a strong focus on handling uncertainty and supporting both distant and close reading. The system is built around a graph that links documents, entity mentions, and entities. Uncertainty from mention-to-entity linking is included directly in this graph, so users can see where connections are strong, weak, or ambiguous. A transfer-function control, inspired by approaches in scientific visualization, allows users to adjust how this uncertainty is displayed, making it easy to tune the visualization for different
datasets and research questions. The system also provides direct access to the full source texts in a coordinated view, enabling quick context checks, resolving ambiguous cases, and correcting digitization errors. Exploration is further supported through a multi-stage filter query builder, mini-map
navigation for large graphs, and export options for downstream analysis. By combining uncertainty-aware graph visualization with direct interaction in the source texts, the system provides a unified workflow that supports both large-scale pattern discovery and curation of named-entity-based document collections. The design choices were supported by the domain experts, who were also involved in the initial system evaluation.
Read the full abstract ▾
Image and Video Processing
CASR: Content-Adaptive Neural Super-Resolution Post-Filter for Versatile Video Coding via Low-Rank Overfitting
Khoa Pham-Dinh, Francesco Cricri, Maria Santamaria et al. (+5 more)
The use of super-resolution as a post-processing step following a
video codec allows
videos to be encoded at reduced spatial resolution at the encoder side and to be reconstructed and upsampled to the original resolution at the decoder side. In this way, the required bitrate is reduced and the quality of the reconstructed frames is improved without modifying the core coding architecture. However, generic SR models are typically trained offline and lack sufficient adaptability to the diverse content characteristics and compression artifacts produced by
video codecs, which limits their effectiveness during test time. To address the limitation, this paper proposes CASR (Content-Adaptive Super-Resolution), a content-adaptive SR post-filter framework for Versatile
Video Coding (VVC), based on encoder-side overfitting on each input sequence. In order to limit the bitrate overhead required for signalling the content adaptation signal, i.e. the weight-update, Low-Rank Adaptation (LoRA) is leveraged. The method freezes the convolution kernels of a pretrained SR network and fine-tunes only lightweight rank-r
matrices attached to selected convolution layers, using VVC decoded frames and quantization-parameter (QP) maps of test sequences as supervision. The resulting low-rank update is compressed with the MPEG Neural Network Compression and Representation (NNR) standard. Experiments on the JVET common test conditions (CTC) class A1 and A2 sequences indicate that LoRA-based content adaptation provides bitrate savings over a non-adapted SR post-filter at a small signalling cost. Compared with the VVC Test Model (VTM21), the proposed method achieves BD-rate savings of -10.93% (Y), -15.39% (U), -24.41% (V) under random access and -13.43% (Y), -5.75% (U), -22.94% (V) under all-intra. An ablation of the LoRA rank r further shows that r=4 provides the best trade-off between coding gain and signalling cost.
Read the full abstract ▾
All authors Khoa Pham-Dinh, Francesco Cricri, Maria Santamaria, Honglei Zhang, Hamed R. Tavakoli, Moncef Gabbouj, Juho Kannala, Miska M. Hannuksela
Signal Processing
Koopman-Predictive Coefficient Tracking for Vehicle-Mounted Monostatic ISAC
Anh Tuyen Le, Xiaojing Huang and Y. Jay Guo
This paper studies predictive coefficient tracking for digitally generated self-interference cancellation (SIC) in vehicle-mounted monostatic integrated sensing and communication (
ISAC) systems. In such systems, the canceller must suppress strong transmitter leakage and nearby unintended reflections while preserving delayed sensing echoes. On mobile platforms, mechanical vibration from engines, rotors, or nearby structures can induce quasi-periodic variation in the self-interference (SI) channel. The optimal cancellation coefficients may then vary within one adaptation period, causing the applied coefficients of a conventional adaptive filter to lag behind the moving SI channel and increasing the residual SI. To address this problem, we propose a Koopman-predictive SIC (KP-SIC) module that augments a conventional adaptive filter with a predicted coefficient trajectory. KP-SIC uses recent effective SI-channel coefficient snapshots to identify structured vibration-induced dynamics and predict a sample-varying coefficient sequence for the next adaptation period. The predicted coefficients place the canceller closer to the expected time-varying optimum, while the adaptive filter corrects residual prediction errors and unstructured variation. For evaluation, the module is integrated with a widely linear normalized least mean-square (WL-NLMS) filter under a flexible analog SIC (FASIC) framework. Simulations show that KP-SIC improves SI suppression by $6.3$–$10.2$~dB over conventional FASIC-WL-NLMS once the observation window spans at least about $1.5$ vibration cycles. Compared with a non-causal performance bound, KP-SIC achieves $60$–$74\%$ of the available prediction gain.
Read the full abstract ▾
Artificial Intelligence
Solving Without Stopping: On-Policy Distillation at Small Scale
Hongyang Li, Yiming Zhu, Xiao Li et al. (+3 more)
On-policy
distillation, where a student learns from a stronger teacher's feedback on its own outputs, is a common way to pass
reasoning to smaller models. We analyze what it transfers at small scale, distilling Qwen3-8B into Qwen3 4B, 1.7B and 0.6B students, in thinking mode (reason at length, then end the
reasoning and answer) and, for comparison, in non-thinking mode (no separate
reasoning phase). Long
reasoning needs two abilities, solving a problem and knowing when it is solved, and we find that
distillation transfers the first, but in thinking mode not the second. Solving improves at every size, up to two ceilings, which we measure comprehensively across both modes and all student sizes: a student's single attempt never exceeds what it could already reach in many attempts before training, and the smaller the student, the further it stays below the teacher. Stopping is where the modes part. In non-thinking mode every student keeps stopping; in thinking mode students stop ending their
reasoning early in training, and the smaller the student, the less of this ability survives: the teacher signals a stop almost only where a student already ends its
reasoning, so
distillation teaches no new stops; it only keeps the student's existing stops that land on a right answer, and a weak student has few such stops. The smallest students often reach the right value but do not commit to it: they either rarely mark it or mark it and write past it. Together, these results describe how small students behave under on-policy
distillation, and a diagnostic that separates answer marking, correctness and stopping.
Read the full abstract ▾
All authors Hongyang Li, Yiming Zhu, Xiao Li, Caesar Wu, Said Mammar, Pascal Bouvry
Lattice Field Theory
Boundary Condition dependent Universality Classes on a Hyperbolic Lattice
Pallabi Dey, Debasish Banerjee, Arnab Kundu et al. (+1 more)
We study the
ferromagnetic Ising model on finite
hyperbolic tessellations of Euclidean AdS$_2$ with open and wired boundary conditions. Because a finite fraction of
spins remains at the boundary as the
lattice grows, these conditions select distinct thermodynamic behaviours. In the $\{5,4\}$ tessellation, Monte-Carlo simulations using efficient worm
algorithms on open boundaries (OBC) yield a transition near $\beta_c J=0.632(3)$, with susceptibility data collapsing under scaling by the total number of
spins $N$. The fitted finite-size exponents, $1/\bar{\nu} \simeq 0.162$ and $\gamma/\bar{\nu} \simeq 0.743$, differ substantially from the mean-field volume scaling. Wired boundaries (WBC), which correlate boundary
spins, instead show a transition around $\beta_c J=0.348(4)$ to an ordered phase consistent with the mean-field exponents. While the mean-field criticality with WBC is consistent with previous studies and with the suppression of independent boundary fluctuations, the OBC exponents hint at the presence of a new universality class. Results from other tessellations support the robustness of the observed scaling with OBC. We discuss a possible route to interpolate between the different boundary conditions. Our findings show that boundary dynamics must be specified when characterizing critical behaviour on
hyperbolic lattices.
Read the full abstract ▾
All authors Pallabi Dey, Debasish Banerjee, Arnab Kundu, Ritam Sinha
Artificial Intelligence
VISTA: Value-Informed Event Appraisal for Multimodal Emotion Conflict
Jiale Dai, Liuxian Ma, Xiaoke Niu et al. (+5 more)
Conflicting emotional cues can be individually valid: a subdued voice may reflect a blocked goal while a smile satisfies a social obligation. Their interpretation depends on what the event means to the person. We introduce VISTA (Value-Informed Semantic Trust Arbitration), a learned seven-field appraisal interface that conditions modality arbitration on concerns, event relations, and expression conditions while retaining a joint-evidence residual. A log-odds decomposition separates
emotion expectation from cue diagnosticity, motivating an interface that lets appraisal change how evidence is interpreted. With a shared Qwen2.5-Omni-7B backbone and matched training examples and steps, VISTA reaches 64.5% conflict accuracy on CA-MER, improving on modality gating by 2.5 percentage points on conflict and 0.2 on consistency. Shuffling appraisal across scenes or removing its decision connection reduces this benefit. A common frozen-backbone probe reaches 0.600 macro CCC for appraisal readout, compared with 0.505 for
emotion-only fine-tuning. Evaluations across five
benchmarks connect recognition under increasing conflict with appraisal readout and downstream decision use. Together, the analyses and experiments support scene-specific appraisal as an intermediate representation that helps interpret conflicting emotional evidence.
Read the full abstract ▾
All authors Jiale Dai, Liuxian Ma, Xiaoke Niu, Wenjing Zhang, Huiying Zhao, Zhaoxiang Liu, Shiguo Lian, Guojie Song
Machine Learning
Parallel Tempering for Diffusion-Based Combinatorial Optimization
Arman Mielke, Uwe Bauknecht, Thilo Strauss et al. (+1 more)
Discrete
diffusion models have emerged as a powerful paradigm for solving combinatorial optimization (CO) problems on graphs by learning to sample high-quality solutions. A common inference-time approach is to generate multiple candidate solutions independently and return the best-performing sample, improving solution quality at the expense of an increase in computational cost. In this work, we introduce PT-Denoise, an inference-time procedure that allows these concurrent denoising trajectories to interact through parallel tempering, without requiring retraining or fine-tuning of the underlying denoiser. Our method assigns a temperature to each
diffusion process and allows processes to swap temperatures based on their relative performance. This dynamically reallocates promising, low-energy trajectories to colder, more concentrated sampling regimes while allowing higher-energy states to escape local minima through randomized exploration. Experiments on canonical graph-structured CO problems show that our approach consistently improves the quality of the best solution found, while only adding minimal computational overhead.
Read the full abstract ▾
All authors Arman Mielke, Uwe Bauknecht, Thilo Strauss, Mathias Niepert
Artificial Intelligence
Mubric: Mutation Testing-Guided Rubric Generation for LLM Evaluation
Jiayuxuan Yang, Jie M. Zhang, Yiling Lou et al. (+1 more)
Rubric-based evaluation is widely used to assess LLM-based systems by decomposing response quality into task-specific scoring criteria. However, automatically generating rubrics that reliably capture task-specific quality requirements remains challenging. We introduce Mubric, a mutation testing-guided approach to rubric generation. Mutation testing, a classic software testing methodology, evaluates a test suite by injecting faults into programs and checking whether the tests detect them. We draw an analogy between test suites and rubrics: if a rubric captures an important quality requirement, introducing a corresponding defect into an otherwise high-quality response should reduce its score. Mubric first mines common defects from real pairs of preferred and dispreferred responses and abstracts these defects into reusable mutation operators, each specifying how to introduce a particular type of response defect. For a new task, it applies relevant operators to a reference response, checks whether the injected defects reduce response quality, and uses insufficiently penalized defects to refine the rubric. We evaluate Mubric on 703 tasks across four representative domains against six advanced rubric generation methods. Mubric achieves the highest overall evaluation accuracy, outperforming the strongest baseline by 7.48 percentage points.
Read the full abstract ▾
All authors Jiayuxuan Yang, Jie M. Zhang, Yiling Lou, Zhenpeng Chen
Multiagent Systems
PowerMarketJax: A JAX Benchmark Suite for Multi-Agent Reinforcement Learning in Power Markets
Zhanhua Pan, Xin Qin, Xiao Liu et al. (+3 more)
Power
markets are a natural testbed for multi-
agent reinforcement learning (MARL), where multiple self-interested participants repeatedly submit bids. A
market-clearing mechanism then determines dispatch and prices subject to power grid constraints and
market settlement rules. However, existing MARL environments typically focus on a single
market setting, implement simplified clearing mechanisms, or rely on CPU-based optimization solvers that slow large-scale training and limit the systematic study of bidding strategies and
market behavior. We introduce PowerMarketJax, a
benchmark suite for MARL across five power
markets: day-ahead wholesale, real-time balancing, ancillary services, peer-to-peer double auctions, and local flexibility. Each environment implements its own clearing, pricing, and settlement rules while providing a common framework for learning and evaluation. We find that learned bidding behavior depends strongly on the
market design: independent learners can miss better strategies when gains require many
agents to change together, when more profitable strategies lie beyond a region of lower profit, or when profits disappear as more
agents adopt the same strategy. PowerMarketJax implements both
market simulation and policy training in JAX, allowing the entire pipeline to run on the GPU with 1,024 X 1,200 parallelisms across both environments and
market participants, achieving up to 33X speedup over CPU-based baselines. Our open-source
benchmark is available at: https://github.com/powermarketjax/PowerMarketJax.
Read the full abstract ▾
All authors Zhanhua Pan, Xin Qin, Xiao Liu, Zhilong Cao, Jianhong Wang, Dawei Qiu
Solar and Stellar Astrophysics
On the origin of carbon- and oxygen-rich hot subdwarfs: an sdB merger scenario
Bo Tang, Xinyu Ji, Songyu Hu et al. (+2 more)
Carbon- and oxygen-rich hot subdwarfs (CO-sdO
stars) represent an extremely rare subclass of helium-rich hot subdwarfs. Recent observations of PG 1654+322, PG 1528+025, and UCAC4 108 reveal unusually large surface mass fractions of carbon and oxygen, which are difficult to explain within existing formation scenarios. We explore whether mergers involving sdB
stars can provide a viable formation channel for these objects. Using the
stellar evolution code MESA, we model the post-merger
evolution of two merger channels: mergers between two sdB
stars and mergers between a helium white
dwarf and an sdB
star. Carbon and oxygen synthesized during core helium burning in sdB
stars may provide a natural chemical reservoir if efficient mixing occurs during the merger. The resulting evolutionary tracks are broadly consistent with the effective temperatures, surface gravities, and luminosities observed for the three
stars, suggesting remnant masses of about 0.8-1.0 M_sun. Our calculations suggest that mergers involving sdB
stars may represent a viable formation pathway for CO-rich hot subdwarfs.
Read the full abstract ▾
All authors Bo Tang, Xinyu Ji, Songyu Hu, Zhongyang Liu, Xianfei Zhang
Computational Complexity
Rational Identity Testing for Noncommutative Circuits is in Polynomial Space
V. Arvind and Pushkar S. Joglekar
The rational identity testing problem, RIT, asks whether an input rational
circuit, computes the zero element of the free skew field. For rational formulas the problem is known to be in deterministic
polynomial time. For rational
circuits of size $s$ the associated linear pencil has dimension $2^{O(s)}$, and the known
algorithms use exponential time and exponential space. We show that RIT for rational
circuits over $Q$ and finite fields is in PSPACE. As a consequence, noncommutative PIT for degree unrestricted noncommutative
circuits is also in PSPACE. The proof is based on the following three observations Analyzing the Hrubes-Wigderson construction of a linear pencil for an input rational formula, we obtain a succinctly represented linear pencil $A$ of size $2^{O(s)}$ for the input rational
circuit of size $s$. More precisely, given indices $i$ and $j$ for the linear pencil we can compute $A_{i,j}$ in space
polynomial in $s$ and length of $i,j$. This
algorithm essentially gives a succinct presentation of the linear pencil $A$ obtained by the reduction. By the recent
theorem of Chatterjee, Ghosh, Gurjar, Raj and Thierauf that deciding whether a symbolic
matrix $\sum_i A_ix_i$ has full noncommutative rank is in NC. We note that their
algorithm is logspace-uniform NC and we use it as a black-box for computing rank of succinctly represented pencil $A$. Finally, we note a folklore simulation: If a problem is solved by a logspace-uniform family of deterministic Boolean
circuits of polylogarithmic depth, and its input is not written down but is presented by a polylogspace subroutine that returns any requested input bit, then the
circuit can be evaluated in polylogarithmic space. Hence, simulating a depth $O(\log^{i}M)$
circuit for succinctly presented inputs of length $M = 2^{\Theta(s)}$ yields a
polynomial space
algorithm.
Read the full abstract ▾
Quantum Physics
Jonathan Paul Home, Jeremy Flannery, Matteo Mazzanti et al. (+2 more)
We outline a complete set of operations for manipulating
nuclear spin qubits of trapped-ions stored in high
magnetic fields such as required for Penning trapping, where the
nuclear and electron
spins are largely decoupled due to the dominance of the external field
Hamiltonian. The reduced
nuclear magnetic moment results in insensitivity to external
magnetic fields compared to the use of an electron
spin, but also makes the
nuclear spin hard to manipulate on fast timescales. To maintain speed, we propose a two-
qubit phase gate technique which utilizes the electron
spin, but for which the
qubit remains protected from
magnetic fields through retaining the
nuclear encoding. This method works for
magnetic-field insensitive
qubits at lower fields as well as for both microwave and
laser field gradients.
Read the full abstract ▾
All authors Jonathan Paul Home, Jeremy Flannery, Matteo Mazzanti, Jan Apolin, Shreyans Jain
Computer Vision
What Comes Next? Omni-StoryBench for Evaluating Story-Grounded Omnimodal Generation
Sieun Hyeon, Yejoon Lee, Mintaek Lim et al. (+3 more)
Omnimodal evaluation should go beyond independent text, image, and
speech production: individually plausible outputs may not express a coherent shared event. We introduce Omni-StoryBench, a story-grounded omnimodal
benchmark evaluating whether models can coherently continue stories across image, narration, and
speech. Each instance provides a current storybook page and structured next-page conditions, requiring models to generate the next illustration, narration, and spoken character utterance. Omni-StoryBench contains 900 rigorously validated story transitions from openly licensed children's books, with ground-truth next-page references and
speech metadata. We evaluate systems with modality-specific metrics and consistency-centered LLM-as-a-judge rubrics for context preservation, condition following, reference consistency, and cross-modal coherence. Across 32 baseline configurations spanning orchestration, semi-orchestration, and native any-to-any paradigms, we find orchestration with strong VLM planning most reliable, while current native omnimodal models often struggle with output completeness and controllability. Our analysis shows text-side performance is associated with image and
speech quality, but image generation and visual continuity form the clearest observed bottleneck among the evaluated configurations. These results position Omni-StoryBench as a system-level
benchmark measuring coherent omnimodal generation beyond isolated modality quality.
Read the full abstract ▾
All authors Sieun Hyeon, Yejoon Lee, Mintaek Lim, Woojin Kim, Jaeik Kim, Jaeyoung Do
Robotics
Battery-Aware Reinforcement Learning for Aggressive Quadrotor Flight
Alejandro Sanchez Roncero, Olov Andersson and Petter Ogren
Agile flight tasks such as drone racing and pursuit-evasion require strong acceleration and precise turns, but the available thrust changes as the battery discharges and voltage drops under load. Conservative command limits make this variation easier to tolerate, at the cost of unused performance. We investigate how learned controllers can use that additional thrust while retaining the flight controller's voltage compensation and rate control. Our training simulator couples an identified load-transient battery model to rotor dynamics and firmware saturation. The feedforward policy receives filtered voltage during both training and deployment. Controlled ablations distinguish the benefit of a larger thrust-command range from that of voltage information. On a 38 g Crazyflie Brushless, the resulting policy reduces circle tracking error by 49% relative to stock-authority RL at 3.84 m/s, while preserving easy-task precision. Mean 20-lap race time decreases from 106.22 s to 95.24 s. Compared with a voltage-blind policy with the same increased authority, hardware error and race time are lower by 15.3% and 4.5%, respectively. In simulation, replacing the policy's voltage input with a recording from a different battery condition worsens hard-circle tracking, with a smaller, voltage-dependent effect in racing. Together, these results show where a simple voltage input complements existing actuator compensation in aggressive learned flight.
Read the full abstract ▾
Software Engineering
Do Agent Benchmarks Do What They Say? An Executable-Contract Audit of Tool-Using Agent Environments
Rohith Reddy Bellibatlu, Zichong Wang and Wenbin Zhang
Tool-using
agents are entering settings where a wrong action carries real cost, and the
benchmarks certifying them grade what each simulated tool call reports having done, assuming the tool did what its interface advertises. The audit taxonomies we survey publish no category for that assumption, and a defect beneath a score is present on every rerun. We treat a tool's advertised surfaces as an executable contract, check the implementation against it, and trace each score's provenance through the task files and evaluator code to the verdicts that derive from state a defective tool should have written. Across 34 audited mutating tools in four
benchmarks we confirm seven tool defects and one evaluator property at pinned commits. On injected defects the checker raised no false positive in 25 flags, flagged 2 of 5 negative controls, and missed most: in 29 of 33 scored misses a clause covered the defect but no probe revealed it. The checker's own static half, run alone, flags 14 of 17 confirmed sites, so on these findings the dynamic half confirms and traces rather than discovers. Twelve further AgentDojo tools, with six held-out tools and the seven audited first, complete its 25-tool mutating surface, on which at least 5 tools diverge from their advertised surface as our contracts read it, a rate for AgentDojo alone. No gold trajectory reaches either tau2-bench defect; on 1,120 paths built to isolate the telecom defect, a number fixed by construction, the evaluator rewards a refuel of a suspended line and fails the repaired tool. The clearest case is a clinical
benchmark whose tool tells the
agent each write executed under a documented no-write design its interface does not disclose; its grader takes that message as evidence, so its action success rate records whether a request carried the expected payload, not whether any record changed.
Read the full abstract ▾
Computer Vision
FLASH: A "Generate Once, Synthesize Many" Framework for Synthetic Anomaly Generation in Industrial Anomaly Detection
Abhay Kumar Das, Rajesh Gangireddy, Ashwin Vaidya et al. (+1 more)
Synthetic anomaly generation helps expand industrial anomaly
datasets when real defects are scarce or unavailable. Existing approaches lie at two extremes: procedural approaches are fast but struggle to represent complex anomalies, while generative approaches produce diverse defects but require costly per-sample generation. We present FLASH, a framework that decouples defect generation from anomaly synthesis under a “generate once, synthesize many” paradigm. Given only normal images, FLASH uses
Vision-
Language Model (VLM) guidance and an image-generation model to produce a small set of defect images, from which it extracts, validates, and banks reusable defect patches. For synthesis of anomalous images, Object Boundary Suppression (OBS) first identifies the probable foreground object-aware region of the host image, while Multi-Resolution Spectral Pyramid (MRSP) noise generates diverse, size-controllable masks that determine the defect location and spatial extent. It then composes a large and diverse synthetic anomalous image set by localizing the defect region, sampling size-controllable placement masks and seamlessly blending retrieved defects onto new defect-free images without further need for image generation. Experiments on the MVTec AD 2
dataset show that FLASH-generated anomalies nearly close the calibration gap on real defects, reaching 78.1% image-level F1 against an 83.6% real-anomaly upper bound and providing the most consistent calibration transfer across detectors among procedural and generative alternatives. Moreover, FLASH synthesizes anomalies more than 11.95x faster than per-sample generative approaches.
Read the full abstract ▾
All authors Abhay Kumar Das, Rajesh Gangireddy, Ashwin Vaidya, Samet Akcay
Analysis of PDEs
David Krejčiřík and Rossano Sannipoli
We investigate the interplay between the geometry of a planar domain and the spectrum of the nonlinear $p$-
Laplacian, subject to a homogeneous
magnetic field and Dirichlet boundary conditions. First, we establish a lower bound for the bottom of the spectrum on the whole plane, extending the classical Landau-level estimate to the nonlinear regime. Second, we prove Pólya-type bounds for the first
eigenvalue on bounded convex planar domains in terms of the area, perimeter, minimal width and
magnetic field strength, and show that they are asymptotically sharp along sequences of suitable thinning domains. Finally, we post some remarks with open problems.
Read the full abstract ▾
Machine Learning
Hidden Reasoning Must Leak, but Need Not Be Readable: Fundamental Opportunities and Limits for Chain-of-Thought Monitoring
Mohammadali Mohammadkhani, Madhava Krishna, Yash Sarrof et al. (+1 more)
Can
reasoning models trick chain of thought (CoT) monitors and perform hidden computation without revealing it in their thinking traces? We show that the answer depends on the underlying task difficulty and the model size. Simple computations can be performed covertly; however, beyond a threshold depending on model size, successfully solving the task necessarily leaks a near-linear amount of information about the covert task input into the CoT. Therefore, sufficiently complex hidden computation always leaves an information-theoretic footprint. However, concerningly, this leakage need not be readable: Under plausible cryptographic assumptions, even a one-layer
Transformer can encrypt its
reasoning online so that no
polynomial-time monitor can extract information about the hidden computation. Overall, our theoretical and empirical results provide a holistic view of both the opportunities and the limitations of CoT monitoring.
Read the full abstract ▾
All authors Mohammadali Mohammadkhani, Madhava Krishna, Yash Sarrof, Michael Hahn
Artificial Intelligence
ReMem: Rethinking Perception and Memory in Long-Context Recommendation Agents
Haohao Qu, Yongcheng Jing, Chun Hin Chan et al. (+3 more)
Recent Recommendation
Agents (RecAgents) offer a promising alternative by shifting recommendation to an active, user-side paradigm, where generative
agents autonomously perceive web platforms, reason over user preferences, and execute decisions. However, existing RecAgents still suffer from two critical limitations: brittle item perception based on noisy item pages, and inefficient long-context
reasoning over extended user histories and multi-step interaction traces. To address these challenges, we propose a recommendation-
agent framework, termed ReMem, that combines OCR-based
multimodal perception with time-evolving dynamic
memory. Instead of parsing raw HTML, ReMem observes item-page screenshots and uses an OCR tool to produce structured textual evidence for the
reasoning backbone. To support long-horizon preference modeling, ReMem further introduces a chunk-wise sequential update strategy that maintains a bounded textual
memory while processing histories longer than the backbone context window; with fixed chunk and
memory budgets, the number of update calls grows linearly with history length. This design allows the
agent to preserve evolving user preferences without relying on external
memory modules or disrupting the standard autoregressive generation process. To train the dynamic
memory updater, we further develop a multi-
memory GRPO variant whose reward combines recommendation-specific information with final-task metrics. Experiments on three Amazon-derived
datasets and a controlled post-
retrieval protocol show an average relative improvement of 7.58% over the strongest reported baseline across searching, ranking, and judging tasks.
Read the full abstract ▾
All authors Haohao Qu, Yongcheng Jing, Chun Hin Chan, Shanru Lin, Wenqi Fan, Dacheng Tao
Cryptography and Security
AutoMark: Enabling Autoresearch to Discover Better LLM Watermarks
Thibaud Gloaguen, Robin Staab and Martin Vechev
With LLM watermarking being deployed commercially and now required by regulations, improving its reliability and effectiveness has become crucial. Yet, recent progress in the field of LLM watermarking has increasingly been driven by improving details of existing methods, an effort fundamentally limited by the pace of human researchers. In this work, we enable for the first time the autonomous discovery of new distortion-free state-of-the-art watermarking schemes. To enable this, we (i) establish strict criteria to ensure that watermarks are reliable (e.g., they do not have an unexpectedly high false positive rate), (ii) propose rigorous statistical tests to automatically evaluate whether a watermarking scheme satisfies our criteria, and (iii) design an evaluation suite to rank watermarks along three key dimensions: detectability, quality, and robustness. By running our framework with 3 frontier models (GPT-6 Astra, Opus 5, Gemini-3.8 Flash), we discover over 50 different watermarking schemes, including several that outperform prior works along all key dimensions. We complement this by a manual study of the discovered schemes, distilling the key ideas into smaller components, and individually studying the impact of each component across dimensions (detectability, quality, robustness) to better understand how the proposed schemes operate. Importantly, we find that the
agents, on top of improving existing ideas, also discover fundamentally new ideas (e.g., aligning watermark scores with random per-request direction). Overall, our work establishes the first steps of fully autonomous watermarking research, enabling the discovery of more reliable and effective watermarks. Our code is available at https://github.com/eth-sri/automark, and a blogpost to visualize our results at https://www.sri.inf.ethz.ch/blog/automark.
Read the full abstract ▾
Differential Geometry
Stable ample vector bundles without semipositive Kobayashi curvature
Kefeng Liu and Xueyuan Wan
We construct slope-stable ample rank-two vector bundles on every Hirzebruch surface $\mathbb{F}_e$ that admit no smooth strongly pseudoconvex Finsler metric with semipositive Kobayashi
curvature. These examples give a negative answer to Kobayashi's problem. The same bundles give counterexamples to the smooth strongly pseudoconvex Finsler formulation of Fulton's problem and to the Hartshorne–Schneider
conjecture on the convexity of complements. For $e=1$, these counterexamples have a simply connected Fano base that is not a product.
Read the full abstract ▾
Numerical Analysis
Hybrid Joint-Selective Optimization: Reduced-Space Levenberg-Marquardt Refinement of Low-Dimensional Parameters of Interest
Muhammad Luthfi Shahab, Gabriella Alfa Indahsari, Imam Mukhlash et al. (+1 more)
This paper introduces a hybrid joint-selective optimization (HJSO) framework for large-scale numerical problems in which a small subset of trainable quantities is of primary interest. We partition the full parameter vector into a high-dimensional remaining block and a low-dimensional block of parameters of interest (POIs), perform joint first-order optimization over the full parameter set, and then freeze the remaining variables while applying a reduced-space Levenberg-Marquardt (LM) refinement to the POIs. The method is designed for settings in which the POIs are low-dimensional but strongly influence the quality of the computed solution, while the full parameter space remains too large for full-space second-order methods. The framework is evaluated on three representative problems: a
matrix eigenvalue problem, an inverse Bratu problem solved with a physics-informed neural network, and a 100-dimensional nonlinear Black-Scholes problem solved with the DeepBSDE method. In each test, HJSO reaches prescribed POI-error thresholds faster than the corresponding joint first-order baseline and improves the final POI accuracy for the reported solver configurations. The contribution is therefore not a universal optimizer, but a practical reduced-space strategy for problems with known low-dimensional parameters of interest and expensive high-dimensional training variables.
Read the full abstract ▾
All authors Muhammad Luthfi Shahab, Gabriella Alfa Indahsari, Imam Mukhlash, Hadi Susanto
Robotics
ActMem-VLA: Action-History Memory with Dual-Expert Denoising for Long-Horizon Vision-Language-Action Policies
Yaxin Zhao, Dianye Huang, Chenwei Wang et al. (+2 more)
Vision-
language-action (VLA) models have driven rapid progress in robotic manipulation, demonstrating strong fine-grained control and promising performance on long-horizon tasks. However, many existing VLAs lack explicit access to interaction history, making them vulnerable to perceptual aliasing: similar current observations and
robot states at different task stages may induce action ambiguity and lower success rate. Existing methods incorporate temporal or progress cues through feature conditioning, action-prior modification, or sampling guidance. However, methods that jointly fine-tune
memory modules and the base VLA incur additional policy-training costs, motivating the separation of trainable history-conditioned steering from frozen base-policy refinement. We propose ActMem-VLA, a dual-expert handover architecture that augments a frozen, fine-tuned VLA with a
memory plugin comprising a Mamba-based
memory module and a lightweight PreAction Expert (PAE). Specifically, Mamba encodes executed-action history into
memory that conditions PAE alongside current context. With these inputs, PAE steers task progression during early, high-noise denoising, then passes the partially denoised action to the frozen Action Expert (AE) to refine action details during the remaining low-noise steps. The fine-tuned base VLA remains frozen throughout training, while only the Mamba module and PAE are jointly optimized. On LIBERO-Mem, ActMem-VLA achieves 80.8% average success across all ten tasks, compared with 65.2% for $\pi_{0.5}$ and 49.5% for MemoryVLA, while introducing only 3.45% additional parameters. Across four real-world tasks, it improves the average success rate over $\pi_{0.5}$ by 28.8%.
Read the full abstract ▾
All authors Yaxin Zhao, Dianye Huang, Chenwei Wang, Chenguang Yang, Zhongliang Jiang