Commutative Algebra
Prime Ideals and Projective Closure in the Boolean Polynomial Semiring
Tristan Bishop, Chris Eppolito, Radford Green et al. (+3 more)
In this paper, we investigate two related questions concerning the Boolean
polynomial semiring in two variables. First, we study
prime ideals in $\mathbb{B}[x,y]$, beginning with a complete classification of those generated by linear
polynomials. Second, motivated by the limitations of ideals in capturing
geometric phenomena over semirings, we introduce a notion of projective closure via congruences. We prove that this projective closure coincides with the topological closure in the projective space of
prime congruences.
Read the full abstract ▾
All authors Tristan Bishop, Chris Eppolito, Radford Green, Jaiung Jun, Alex Norwood, Daisy Ivy Thackrah
Statistical Methodology
Adaptive mixture variational inference for spike-and-slab regression
Hanqing Li, Yaroslav Golub and Xuewen Lu
Correlated predictors can support competing sparse explanations with similar predictions, making joint uncertainty about variable inclusion difficult to capture with mean-field approximations. We develop an adaptive fitting procedure for mixtures of product distributions in Gaussian regression with a point-mass spike-and-slab prior. It minimizes reverse Kullback-Leibler divergence directly on inclusion indicators and active coefficients, jointly refining component parameters and weights as the mixture grows. This avoids an additional divergence penalty on unused latent coefficients under independent augmentation. Our analysis relates approximation accuracy to mixture size, support coverage and dependence within supports, and establishes contraction, selection consistency and a Bernstein-von Mises approximation under explicit conditions on the prior, posterior concentration and variational error. On all 250 simulated
datasets with exact posterior references, mixtures reduce errors in inclusion probabilities, grouped support probabilities and coefficient covariance relative to multistart mean field. Comparisons at fixed mixture size and common initialization favor direct joint refinement over augmented or restricted refinement in posterior divergence. Complete stagewise fitting can nevertheless be more accurate near collinearity. The results support direct joint refinement for posterior approximation while showing that local gains do not ensure superiority of the full adaptive search.
Read the full abstract ▾
Systems and Control
Forward-Invariant Policy Classes for Safe Reinforcement Learning in Multicopter Control
Chieh Tsai, Muhammad Junayed Hasan Zahed, Jinzhi Shen et al. (+5 more)
This paper proposes a reinforcement learning (RL) framework for safe gain scheduling based on forwardinvariance-induced action-space design. Under stated nominalmodel and inversion-domain assumptions, rather than enforcing safety through runtime shielding or penalty-based constraints, safety is embedded directly into the policy class. Specifically, we construct a finite library of feedback controllers sharing a common Lyapunov certificate that establishes forward invariance of a prescribed admissible set under arbitrary switching. Consequently, any policy whose actions are restricted to this library, including policies encountered during RL exploration, inherits the same certificate. Policy optimization can therefore focus on closed-loop performance without runtime safety filtering or action projection. The framework is instantiated for quadcopter hover regulation, where a DQN schedules among certified feedback controllers. Nonlinear MuJoCo simulations demonstrate state-dependent gain scheduling and empirically evaluate robustness to wind, model mismatch, sensor noise, and sensing delay. The results illustrate how safety certification can be separated from policy optimization by learning over a forward-invariant policy class.
Read the full abstract ▾
All authors Chieh Tsai, Muhammad Junayed Hasan Zahed, Jinzhi Shen, Yi Xie, Ruoshan Lan, Majed Obaid, Salim Hariri, Hossein Rastgoftar
Fluid Dynamics
Directional Control of Droplet Motion via Geometric Gating
Manoj K. Chaudhury, Vanessa Elias and Michelle Fruge
We explore new methods for inducing the movement of water droplets on elastomeric substrates, such as polydimethylsiloxane (PDMS). Our approach builds upon the well-known phenomenon that water droplets move easily across these substrates when swelled w,ith silicone oil. While designing shallow open channels and traps on a PDMS film offers a simple way to control a droplet's trajectory by tilting the substrate in two orthogonal directions, this method is poorly suited for manipulating multiple droplets simultaneously. To address this limitation, we introduce two new methods. The first method utilizes a plano-concave PDMS platform with a radius of
curvature significantly larger than the size of the droplet and its translation distance. When the system's configuration is altered, this curved substrate consistently guides the droplet along a curved geodesic. The second method incorporates diode-like gates on the PDMS. These gates are created on either flat or plano-concave PDMS films by
embedding thin, rigid plates into the elastomer prior to crosslinking, followed by swelling in silicone oil. This process generates sigmoidal steps that restrict droplet motion to a single direction. Finally, we demonstrate how combining these two strategies enables various forms of directed motion for both single and multiple droplets.
Read the full abstract ▾
Robotics
TERRA: Terrain-Aware Reconstruction, Retargeting and Control for Musculoskeletal Locomotion
Merkourios Simos, Chengkun Li, Bianca Ziliotto et al. (+1 more)
Recent advances in musculoskeletal modeling and reinforcement learning have enabled muscle-actuated
agents to reproduce increasingly complex human motions. Yet these capabilities remain largely confined to flat ground, in part because motion
datasets rarely include aligned terrain geometry and because retargeting terrain interactions to complex musculoskeletal bodies is challenging. We present TERRA, an end-to-end pipeline for terrain-aware retargeting and control of musculoskeletal locomotion. From kinematic trajectories alone, TERRA combines terrain priors, estimated contacts, and negative free-space evidence to recover task-relevant support geometry. TERRA further considers anatomical, tendon-continuity, and contact constraints during retargeting. Using the resulting motion-terrain pairs from five
datasets, we successfully train a single muscle-actuated control policy on 9.4 hours of diverse locomotion. Across reconstruction, retargeting, and held-out tracking
benchmarks, TERRA improves terrain accuracy, sharply reduces anatomical and interaction violations, and achieves the highest observed completion rate over supported terrain families. Overall, TERRA provides a practical route from scene-less motion data to muscle-actuated locomotion over diverse non-flat terrain. Project website: https://cnai.epfl.ch/terra/
Read the full abstract ▾
All authors Merkourios Simos, Chengkun Li, Bianca Ziliotto, Alexander Mathis
Artificial Intelligence
AgBench: Agentic AI Benchmarks for Personal AI Devices
Yizhou Han, Di Wu, Dhananjay Saikumar et al. (+1 more)
Agentic AI systems increasingly rely on cloud-hosted large
language models for planning, tool use, and iterative execution, raising concerns about API cost and data exposure. Advances in personal AI devices enable
agents to execute locally, but limited resources on device may affect task success and performance. Existing
benchmarks are inadequate for systematically characterizing these trade-offs across devices, workloads, and deployment architectures. We present AgBench, a
benchmark suite and open artifacts for reproducible evaluation of agentic AI on personal devices. Using AgBench, we evaluate local, hybrid, and cloud execution across agentic workloads, examining task success, latency, cloud API cost, and data exposure. Our results, drawn from over 162.07 million data points, show that personal AI devices can complete many
agent tasks locally, but local-only execution generally has lower task success and longer completion times than cloud-only execution, especially as concurrency increases. Local-only execution eliminates cloud model API costs and sensitive-information exposure to cloud
agents. Hybrid execution can improve task success, but its cloud cost and data exposure depend on how
agents divide work and share information. No single architecture performs best across task success, goodput, cloud cost, and data exposure; deployment choices should reflect the intended workload and device capabilities. AgBench is available at https://anonymous.4open.science/r/AgBench-2777.
Read the full abstract ▾
All authors Yizhou Han, Di Wu, Dhananjay Saikumar, Blesson Varghese
Genomics
Fragment Path Complex Networks for Viral Genome Classification
Alice Wachira, Xiang Liu and Guo-Wei Wei
Accurate classification of viral families is essential to understanding genomic diversity, but rapid
evolution and heterogeneous genome structure complicate reliable assignment. Newly discovered viruses, particularly those from underrepresented ecosystems, expand the known diversity of viral genomes. This growing diversity poses challenges for classification methods that rely on similarity to existing references. We present Fragment Path Complex Networks (FPCN), a framework that represents viral genomes as sequences of ordered fragments at various scales. FPCN captures contextual sequence information within each fragment and uses path complexes to model complex relationships between neighboring fragments. We evaluated FPCN on well-established NCBI
benchmarks, and FPCN achieved higher classification accuracy than competing methods. Further evaluations on newly released
datasets revealed FPCN's ability to generalize to genomes that were not presented in the training. Together, these results position FPCN as an effective tool for classifying viral genomes within known families across
benchmark datasets and newly released genomes.
Read the full abstract ▾
General Mathematics
Finiteness in Square Classes and Weighted Zero Density of Perfect Cuboids
Valery Asiryan
Let $M(N)$ be the number of similarity classes of positive rational perfect cuboids with ordered edges $a,b,c$, space diagonal $g$, and squarefree invariant $N=\operatorname{sf}(abcg/2)$. We prove that $M(N)$ is finite for every positive squarefree $N$, without any restriction on the rank of the associated congruent-number
elliptic curve. More precisely, $M(N)\le C^{1+r_N}$ for an absolute constant $C>1$, where $r_N$ is the rank of $y^2=x^3-N^2x$. Using the established Paulsen-West rank obstruction and Smith's results on quadratic twists, we then prove \[ \lim_{X\to\infty}\frac1X \sum_{\substack{N\le X\\N\text{ positive squarefree}}}M(N)^q=0 \qquad\text{for every fixed }q>0. \] Thus the result counts cuboid classes with their full multiplicities at each $N$. A quantitative decay estimate is also obtained. The
geometric step excludes every positive-dimensional real abelian coset through the positive part of the compatibility surface; quantitative Mordell-Lang then gives the required bound. The argument is unconditional and uses the cuboid correspondence and rank obstruction as prior results.
Read the full abstract ▾
General Relativity and Quantum Cosmology
Tidal Response Beyond the Background Geometry in Einstein-Euler-Heisenberg Black Holes
Ramon Becar, P. A. Gonzalez, Ali Ovgun et al. (+2 more)
We determine the quadrupolar logarithmic
tidal response of electrically charged Einstein-Euler-Heisenberg black holes from the coupled static odd-parity gravitational and electromagnetic equations. Although the second electromagnetic invariant vanishes on the electric background, its quadratic interaction contributes to the axial fluctuation equations. At first order in the Euler-Heisenberg coupling $a$, a direct asymptotic recursion gives the canonical running
matrix $C_{\rm can}^{\rm EH}=(9aQ^2/10)\left(\begin{smallmatrix}0&-Q\\-Q&M\end{smallmatrix}\right)$, defined as the coefficients multiplying the canonical source vector in the $r^{-2}\log(r_0/r)$ term. The resulting local coefficients retain the full dependence on the subextremal ratio $\frac{Q}{M}$ at this perturbative order. The two quartic invariants partially cancel in both nonzero channels, while the action-normalized fields make gravitoelectromagnetic reciprocity explicit. The local extremal limit of the running
matrix reproduces the corresponding two-invariant effective-field-theory result after convention conversion. Allowing independent quartic couplings further shows that identical electric background geometries can have different axial running
matrices and yields algebraic inverse relations within the specified two-operator model. The calculation determines local logarithmic running within minimally coupled quartic electrodynamics; the finite horizon-matched response requires an additional global matching calculation.
Read the full abstract ▾
All authors Ramon Becar, P. A. Gonzalez, Ali Ovgun, Joel Saavedra, Yerko Vasquez
Operating Systems
StateFork: Branchable Infrastructure for Agent Exploration
Jiakai Xu, Tianle Zhou, Georgios Liargkovas et al. (+5 more)
AI
agents improve task success by exploring multiple trajectories, but for computer-use
agents each trajectory modifies external environment state. Branching from an intermediate point is correct only when restoration is observation-equivalent - future actions produce the same observations - and practical only when creating, restoring, and discarding branch states is physically efficient. We study this problem for terminal-using
agents, where tasks modify files, shell context, running processes, and local services. We introduce StateFork, a logical control plane that separates exploration policies from physical state materialization, exposing sessions, commands, snapshots, restores, and cleanup over multiple execution substrates. We also build Waypoint, a checkpoint/restore substrate for terminal execution sessions that combines filesystem layering, process checkpointing, and a persistent terminal-compatible command session. Together, StateFork and Waypoint improve terminal-
agent exploration by combining sample-efficient search with efficient restoration of the right execution state. On Terminal-Bench, branch-based exploration through StateFork and Waypoint improves task completion over pass@20 at the same visited-node budget, and using Waypoint achieves 26% higher task accuracy than other execution substrates while completing exploration up to 70% faster. These results show that observation-equivalent, physically efficient execution sessions are a key systems abstraction for exploratory AI
agents.
Read the full abstract ▾
All authors Jiakai Xu, Tianle Zhou, Georgios Liargkovas, Danielle Gillai, Ruizhe Fu, Patrick Shen, Eugene Wu, Kostis Kaffes
Machine Learning
Uncertainty-Normalized Margins for Direct Preference Optimization
Sadegh Khorasani, Petrus Mikkola and Matthias Grossglauser
Direct preference optimization (DPO) models binary preferences through a Bradley-Terry model with a common noise scale, without explicitly accounting for preference strength or
prompt-dependent uncertainty from human feedback. We introduce uncertainty-normalized margin DPO (UNM-DPO), which combines strength-dependent margins with a learned
prompt scale. Motivated by a heteroskedastic Bradley-Terry model, we develop two training objectives. Both compare the implicit rewards of preferred and rejected responses, derived from response log-probability ratios to a reference policy. Advantage-only (AO) divides this reward difference by the
prompt scale before subtracting the margin; whole-residual (WR) subtracts the margin before dividing by the scale. For the WR comparison model, we establish a necessary and sufficient condition under which known margins make the
prompt scale identifiable. We introduce a practical procedure for learning the scale. Building on WR, we introduce ULNM-DPO-WR, which normalizes each response's implicit reward by its length. We evaluate our methods against DPO and related baselines on HelpSteer2 and HelpSteer3, using the Skywork reward model as a judge. With Llama-3.1-8B-Instruct, ULNM-DPO-WR achieves tie-adjusted win rates against matched DPO of 68.00% and 65.31% on evaluation panels, with higher mean rewards and shorter responses on average. On AlpacaEval with a GPT-4.1 judge and GPT-4-Turbo reference answers, the same 8B policy achieves a length-controlled win rate of 21.62%, compared with 16.39% for DPO and 15.30% for SimPO. These results demonstrate the potential of combining preference-strength margins, learned
prompt scales, and length normalization for policy optimization.
Read the full abstract ▾
Information Retrieval
Exploring Forum Post Retrieval with Generative Modeling
Yang Li, Yaguang Liu, Heng Liu et al. (+14 more)
Generative recommendation (GR) has emerged as an alternative to
embedding-based
retrieval, building on the success of generative models in
language and
vision. We are exploring GR on Facebook Forum, a standalone application for medium-to-heavy users of Facebook Groups. Because Forum is a new surface, its own interaction data are too sparse to train a GR model from scratch. We address this with transfer along two axes: we train on a broader corpus of Facebook Groups engagements rather than Forum sessions alone, and we reuse hierarchical, prefix-based semantic IDs (SIDs) learned from cross-platform Facebook Feed data instead of fitting a Forum-specific tokenizer. A 3B-parameter instruction-tuned
language model is then supervised-fine-tuned to generate SIDs directly from user context. We systematically ablate the design choices that matter most in practice, including SID construction, the composition and length of user history, and the inclusion of user-profile features. Our results show that cross-platform SIDs transfer to a new recommendation surface, and offer practical guidance for teams deploying GR on real-world social platforms.
Read the full abstract ▾
All authors Yang Li, Yaguang Liu, Heng Liu, Samson Komo, Jane Kou, Yulian Zhou, Gang Yang, Shubhojeet Sarkar, Gaurav Chakravorty, Yujie Liu, Haipeng Chen, Yonghuan Yang, Deepti Chheda, Yamin Wang, Mike Plumpe, Rish Tandon, Shengbo Guo
Machine Learning
Alignment via Training Against Probes Without Losing Monitorability
Lena Libon, Alexander Panfilov, Ben Rank et al. (+3 more)
Models are usually aligned based on their observed outputs, using demonstrations, preference data, or reward signals. These objectives reward responses that look aligned. More capable models may learn to satisfy them without internalizing the intended behavior, for example by faking compliance during training. Such superficial compliance could be harder when the objective is defined on model internals rather than outputs. Therefore, we study probe-guided fine-tuning, using probes that detect undesired properties in model activations as a direct training signal. We evaluate linear and non-linear probes with different numbers of probes per layer across two
alignment objectives: harmlessness and honesty. We find that training against probes that do not update during training is an easily exploitable objective, while continuously updated probes substantially reduce harmfulness and improve honesty while preserving utility. Probe-guided fine-tuning achieves better safety-utility trade-offs than DPO and inference-time steering, while being substantially more robust against jailbreak and abliteration attacks. Moreover, the concepts stay linearly encoded after fine-tuning, meaning oversight is not lost by our method. Training against probes thus offers a way to shape what models represent rather than only what they output, which may become increasingly important as models get better at making their outputs look aligned.
Read the full abstract ▾
All authors Lena Libon, Alexander Panfilov, Ben Rank, Xin Chen, Jonas Geiping, Maksym Andriushchenko
Image and Video Processing
Joint Supervised and Self-Supervised Training with Acquisition-Robust Techniques for Accelerated 4D Flow MRI Reconstruction
Mengyuan Xue and Bochun Mei
4D flow MRI measures time-resolved, three-directional blood velocity but requires long acquisition times, and its diagnostic signal is carried by the phase difference between velocity encodings, not by image magnitude. Recent work has developed a per-encoding variational network to address image reconstruction in this field. In this work, we incorporate a joint supervised and self-supervised training regime and utilize both magnitude and velocity data during supervision. At the same time, we add multiple acquisition-robust and conditioning strategies based on the acceleration factors. On the CMRx4DFlow~2026 aortic
dataset (1.5 and 3T), our model lowers RelErr by $38$–$50\%$ and AngErr by $7.0$–$8.7^\circ$ against a training-matched baseline across $R{=}10$–$50$, improving on every held-out subject at every acceleration. Our model also shows strong generalization ability to transfer on out-of-distribution data by employing the joint training scheme, with an increase of $7.2\%$ in SSIM and decrease of $38\%$ and $31\%$ in AngErr and RelErr respectively.
Read the full abstract ▾
Multiagent Systems
Recursive Organization Improvement: A Modeling Specification for Human–Agent Organizations
Zilong Wang
Stronger AI
agents do not automatically produce better organizations: teams must also learn which work arrangements to retain and when to reconsider them, while those arrangements determine what evidence the team can learn from. We propose a modeling specification for recursive organization improvement and evaluate it through an executable checker, a public-record mapping, and two simulation studies. The specification connects actor-visible histories, organizational
memory, decision rights, and the evidence used to retain changes. Study 1 crosses six decision rules, three
memory conditions, and three task environments under fixed resource ceilings. In a stationary environment, cumulative evidence raises balanced evaluation's net value per task from 0.452 to 0.480, while selection regret falls from 0.029 to 0.002. After a workflow-ranking reversal, retained evidence delays adaptation. In post hoc acquisition-matched controls, the remaining contribution of program replacement depended on the
memory rule: small and uncertain under an eight-round window, positive under cumulative
memory after a reversal, and negative under reset evidence. Study 2 models human-
agent review with shared errors, copying, and decision rights. It identifies a two-actor case in which public records conceal whether agreement comes from shared errors or copying, and shows how dependence changes the preferred placement of human judgment under a fixed budget. Paid observation can lower decision error while costing more than the deployment gain. Assessments of organizational improvement should track the evidence acquired and retained, using acquisition-matched comparisons to isolate the contribution of evaluator revision.
Read the full abstract ▾
Artificial Intelligence
ChartRevise: A Dataset and Evaluation Protocol for Exact Chart Editing via Code
Jiaxiang Tang, Yi Zhou, Chad DeLuca et al. (+5 more)
Chart editing requires cross-modal edit grounding, realizing a requested visual change in the code that draws it, with necessary related updates and without altering unrelated content. Existing
benchmarks emphasize either code executability or chart quality, but their metrics do not clearly distinguish request completion from missed coupled updates and gratuitous changes. We introduce ChartRevise, a structured
dataset and evaluation protocol for exact program-grounded chart editing. For
dataset construction, we build on the grammar of graphics to systematically cover chart-editing operations, using source-program checks to verify their applicability across chart types and libraries. To improve edit exactness, our pipeline checks individual requirements and guides repair or exclusion when they are unmet. The resulting
dataset contains 92,438 records covering 344 edit types across 20 chart types and three plotting libraries. For evaluation, our reference-free protocol separately measures atomic requirement completion, identifies gratuitous changes, and detects missed coupled updates. These checks are combined with successful execution and rendering to determine exact-edit success. Across five models and four external
benchmarks, fine-tuning yields relative gains of 16% in mean requirement recall and 22% in mean exact-edit rate.
Read the full abstract ▾
All authors Jiaxiang Tang, Yi Zhou, Chad DeLuca, Rogerio Feris, Ahmed Khalil Omran, Zhi-Li Zhang, Pengyuan Li, Ali Anwar
Computer Vision
Kai Yan, Xiangyu Chen, Yulong Cao et al. (+9 more)
Vision-
Language-Action (VLA) foundation models have recently emerged as one of the prevailing solutions for autonomous driving, as they can utilize knowledge acquired during
vision-
language pretraining for accurate and interpretable driving. However, VLAs can take only a limited number of frames as visual input due to the high
token cost of an image, which is problematic for
memory-dependent tasks such as determining the arrival order at all-way stops and long-horizon driving scene understanding. Existing solutions use latent vector
memories accessed through cross-
attention, which are neither interpretable nor portable. In this paper, we propose AD-Memo, a general-purpose VLA driving
agent with
language-based
memory. The
agent outputs
memory as an extension of its Chain-of-Thought (CoT) to record surrounding objects critical to driving; this
memory becomes part of the
agent's future input. We curate
memory-based
datasets and train VLAs with a two-stage recipe: Supervised Fine-Tuning (SFT) and Da Capo, a novel semi-closed-loop Reinforcement Learning (RL)
algorithm which uses trajectory-level advantage for
memory and step-level advantage for driving, leading to better credit assignment. Across scenarios such as all-way stops and general driving, AD-Memo improves driving quality, enables better question answering on driving scenes, and produces plug-and-play
memory for other models.
Read the full abstract ▾
All authors Kai Yan, Xiangyu Chen, Yulong Cao, Alex Naumann, Peter Karkus, Yan Wang, Jef Packer, Alex Schwing, Yuxiong Wang, Boris Ivanovic, Wenjie Luo, Marco Pavone
Robotics
Yggdrasil: a Layer-First 3D Scene Graph for Real-Time Querying
Arshia Akhavan, Ermanno Bartoli, Afnan Algharbi et al. (+3 more)
Robotic
agents use 3D scene graphs (3DSG) to perform tasks ranging from scene understanding to scene interaction. Although an extensive body of work addresses scene graph generation, little
attention has been paid to optimizing the graph for consumption, which leaves state-of-the-art perception pipelines to work around their own scene graph and to pay a latency cost that does not fit the real-time budget a perception loop runs on. We present Yggdrasil, the first 3D scene graph designed to be efficient for both generation and consumption: a layer-first hierarchical graph built from generic nodes, edges, and layers, which expresses the representations existing pipelines already produce, indoor or outdoor, flat or hierarchical, while natively answering the positional and semantic queries downstream tasks issue. Against a published DSG baseline, Yggdrasil answers queries up to $121\times$ faster, and every query we measure falls between 2 and 127 microseconds, three to five orders of magnitude inside the 200 microsecond keyframe budget a 3DSG consumer lives in, on both a workstation and embedded class device. We integrate Yggdrasil into three published pipelines spanning human trajectory prediction, object-goal
navigation, and human-aware motion planning, where it removes up to 99% of the time each spends on its scene graph. The implementation,
benchmark harness, and all three integrations are available online.
Read the full abstract ▾
All authors Arshia Akhavan, Ermanno Bartoli, Afnan Algharbi, Alireza Hoseinpur, Iolanda Leite, Bryan Donyanavard
Artificial Intelligence
Component-Aware Feedback for Self-Evolving Programs
Ethan Lin, Jinming Nian and Yi Fang
LLM-guided evolutionary search can discover complex programs, but existing methods mostly only save candidate programs and fitness scores while discarding which component edits produced which fitness metric changes. Existing methods force the mutator LLM to infer the effect of prior edits from cluttered histories, making program search slow and unstable. This is especially true for locally servable LLMs to evolve multi-component systems. We introduce component-aware feedback, which compares each evaluated program with its parent, identifies the components that changed, and logs them with the associated metric differences into an attribution
memory that later mutations read. The
memory keeps each change in two reference frames, local against the parent it came from and global against the seed program, which shows both the immediate effect of a change and the cumulative progress made since the seed. We study this on LLM reranking, a multi-objective optimization problem where a multi-stage pipeline must balance quality against serving cost. Across twelve Bright
datasets, our method reaches the strongest baseline's final quality after a median of one third of the search budget and ends 7.2% higher in held-out nDCG@10, and under a cost-aware objective it finds pipelines that are on average more accurate while using 11% fewer
tokens per query, showing component-aware feedback to be a promising direction for more efficient self-evolving systems.
Read the full abstract ▾
Computational Engineering
SHIFT-Truck: A High-Fidelity Aerodynamics Dataset and Benchmark for Pickup Trucks
Riddhiman Raut, Yin Yu, Aashwin Anand Mishra et al. (+4 more)
Pickup trucks account for 14% of new light-duty vehicles produced in the United States, yet are among the least aerodynamic. Their open cargo bed adds a flow absent from existing automotive aerodynamics
datasets such as DrivAerML and SHIFT-SUV: the shear layer leaving the cab roof passes over a recirculating bed flow before separating again at the tailgate. The resulting drag lowers fuel efficiency, raises emissions and limits the range of electric trucks. Scale-resolved Computational Fluid Dynamics (CFD) is too costly for broad design exploration; neural surrogates can predict flow features at a fraction of that cost, provided they are trained on large-scale, high-fidelity, domain-specific data. We introduce SHIFT-Truck, the first such
dataset for pickup trucks. It comprises 1,000 Spalart-Allmaras delayed detached-eddy simulations (SA-DDES) of a reference pickup geometry morphed across 17 shape parameters. Each case is run on a mesh of about 100 million
cells at a Reynolds number of $1.4 \times 10^7$ and released with time-averaged surface pressure, wall shear stress, volumetric pressure and velocity. The setup is verified by grid refinement and repeated runs, and checked against wind-tunnel measurements. We define geometry-grouped splits and
benchmark four neural surrogates, DoMINO, GeoTransolver, AB-UPT and SMART, on surface and volume tracks. SHIFT-Truck also introduces controlled distribution shifts in the operating point, the input surface discretization and the vehicle archetype. Models with strong in-distribution performance can degrade substantially under these shifts: operating-condition changes expose failures to infer speed dependence, while tessellation and cross-vehicle shifts reveal markedly different robustness across architectures. SHIFT-Truck is thus a
benchmark not only for surrogate accuracy but also for generalization across physical and numerical distributions.
Read the full abstract ▾
All authors Riddhiman Raut, Yin Yu, Aashwin Anand Mishra, Michael Emory, Thomas Economon, Peter Lyu, Juan J. Alonso
Computer Vision
Template-Search Domain Adaptation via Multi-Stage Feature Alignment for Cross-Modal Object Tracking
Fereshteh Aghaee Meibodi, Amir Mehdi Soufi Enayati, Shadi Alijani et al. (+1 more)
Visual object tracking typically assumes that the initial template and subsequent search frames share the same sensing modality. In practice, sensor availability or operation may change over time, creating a substantial representation gap between template and search frames. Unlike conventional multi-modal tracking where paired modalities are simultaneously available, cross-modal tracking requires localization when template and search frames originate from different active modalities. Accordingly, we introduce TSDA-Track, a Template-Search Domain Adaptation framework to reduce modality discrepancy during training. We investigate two feature
alignment strategies. Pre-AFA TSDA-Track applies
adversarial alignment before
transformer's template-search interaction to suppress modality-specific bias. Enc-CFA TSDA-Track applies contrastive
alignment to encoder representations after interaction to strengthen target-level cross-modal correspondence. Both variants retain a shared inference pipeline without modality-specific branches. Experiments on LasHeR, and zero-shot evaluations on RGBT234 and GTOT under multiple cross-modal protocols demonstrate improvements over representative state-of-the-art trackers. For instance, under the modality-switch protocol on RGBT234, Pre-AFA TSDA-Track achieves an SR/PR of 43.2/56.0, compared with 36.8/50.0 for ToMP-101 baseline. In addition, a study on Anti-
UAV-024 further verifies the applicability of TSDA-Track to aerial tracking. Our study highlights the effectiveness of feature
alignment domain adaptation for cross-modal tracking.
Read the full abstract ▾
All authors Fereshteh Aghaee Meibodi, Amir Mehdi Soufi Enayati, Shadi Alijani, Homayoun Najjaran
Computer Vision
STEPS: Scene Text Editing with Preserved Style Using Diffusion and Contrastive Style Encoding
Nicolas Thiebaut, Nameer Hirschkind, Xiao Yu et al. (+1 more)
We introduce Scene Text Editing with Preserved Style (STEPS), a novel
diffusion model architecture for quality text replacement in images. Scene Text Editing (STE), also known as Visual Text Editing, consists of changing the textual content in an image while conserving the original style, e.g. font, colors, orientation, background, etc. STEPS advances the state of the art in STE through directed focus on improved style preservation. We introduce a style encoder for visual text that captures style independently of textual content, and a model architecture that combines the style encoder with multiple semantic conditions (target text characters encoding and rendered glyphs). STEPS achieves superior results to previous STE methods in style preservation, output readability, and subjective quality.
Read the full abstract ▾
All authors Nicolas Thiebaut, Nameer Hirschkind, Xiao Yu, Kyle Spence
Image and Video Processing
TSGL: Teacher-Student Graph Learning for 3DGS Compression
Matin Bani Saedi, Matthew Kyan and Gene Cheung
3D Gaussian Splatting (3DGS) is a popular representation for novel view synthesis. However, 3DGS contains millions of Gaussian primitives, each with rich attributes, resulting in large file sizes. We propose a novel 3DGS compression method based on Teacher-Student Graph Learning (TSGL) that operates directly on a trained model, without 3DGS retraining or access to training images. Specifically, for each block of Gaussian primitives, using decoded positions and DC spherical harmonic (SH) coefficients as predictors, we learn a signal-dependent geometry graph G encoding the pairwise similarities between neighbouring Gaussians via a teacher-student model. Given G, we perform Graph
Fourier Transform (GFT) on the remaining attributes, so that signal energies are predominantly projected into the low-frequency coefficients for compact representation. On three standard
benchmarks, the method reaches 27x to 33x compression with less than 0.6 dB of PSNR loss, improving on recent post-training compression methods in both size and rendering quality.
Read the full abstract ▾
Signal Processing
Estimates of neural health using cochlear implant digital twins correlate with speech recognition
Erin L. Bratu, Andrea J. DeFreese, Katelyn A. Berg et al. (+4 more)
The cochlear implant is a neural prosthetic that uses an array of electrodes to directly stimulate auditory nerve fibers with electrical impulses. One factor that is known to impact outcomes is the quality of the electro-neural interface. In this work, we propose a pipeline that permits creating accurate and patient-specific digital twins that model the cochlear implant electro-neural interface. We leverage medical images and electrophysiological measurements from the patient's device for digital twin construction. We validate our digital twins by comparing predictions of neural response measurements to clinical measurements and
speech recognition outcomes. Neural health estimated by the digital twin correlates strongly with word recognition rates (R=0.75, p<1e-5) in an analysis of 29 cases and sentence in noise recognition rates (R=0.78, p<1e-4) in an analysis of 21 cases. We also report preliminary findings with digital twin-based patient remapping with 6 subjects, where word recognition rates improved on average by 10 percentage points. These digital twins have the potential to provide clinicians with an unprecedented window into the electro-neural interface for individual cochlear implant recipients, which could lead to improved hearing outcomes when used as a foundation for program customization.
Read the full abstract ▾
All authors Erin L. Bratu, Andrea J. DeFreese, Katelyn A. Berg, Margaret Wilson, Aaron C. Moberly, Robert F. Labadie, Jack H. Noble
Quantum Physics
Towards noble gas quantum optical magnetometry using direct ultraviolet detection
James Maldaner and Gil Porat
We analyze a novel approach for
quantum magnetic sensing via optical detection of
nuclear spin precession in a noble gas. The detection is carried out with an ultraviolet frequency comb
laser, so no alkali vapor is required to be mixed with the noble gas during the sensing stage, allowing for hours-long
spin relaxation time at room temperature. Our analysis reveals that combining high
laser power and low
laser intensity noise is crucial to achieving and exceeding state-of-the-art magnetometric performance, including that of cryogenic magnetometers. We find operational regimes where either fundamental or technical noise dominate, which directs experimental effort towards increasing
laser power or lowering
laser intensity noise, respectively. We also show that optimizing beam size is key to achieving optimal sensitivity. Our calculations predict that a record-breaking sensitivity of $24~\mathrm{aT}/\sqrt{\mathrm{Hz}}$ is achievable in 10 minutes with a 10 W ultraviolet
laser with a relative intensity noise of -100 dBc/Hz.
Read the full abstract ▾
Machine Learning
Proper Scoring Rule-based Diffusion for Probabilistic Weather Forecasting
Joonhyeong Park, Giung Nam, Hyungi Lee et al. (+3 more)
Recent probabilistic weather forecasters train
stochastic predictors with the continuous ranked probability score (CRPS) to generate each ensemble member in a single forward pass. These models learn the predictive distribution from the forecast context alone, which becomes difficult at longer forecast horizons where uncertainty is high. To learn the predictive distribution more effectively, we introduce auxiliary conditional denoising tasks that predict the same future state from the context and its corrupted version, which provides partial future information that can reduce prediction ambiguity. Building on distributional
diffusion models, we learn the conditional distributions of these tasks with a single
stochastic predictor by minimizing a proper scoring rule across noise levels. At inference, the predictor can still generate each ensemble member in a single forward pass at the fully corrupted endpoint. Standard CRPS training is recovered as the endpoint-only special case of our formulation, so our framework extends existing CRPS-based forecasters with only additional conditioning inputs. Controlled experiments show that the auxiliary tasks improve one-step
forecasting across architectures, with larger gains at longer forecast horizons. The gains extend to high-dimensional global weather
forecasting under both training from scratch and fine-tuning, along with improved calibration and potential benefits for generalization under distribution shift.
Read the full abstract ▾
All authors Joonhyeong Park, Giung Nam, Hyungi Lee, Kyunghyun Cho, Byoungwoo Park, Juho Lee
Quantum Physics
Passive realism in the presence of open system dynamics
James Fullwood, Boyu Yang and Weixiang Ye
We introduce the terminology passive realism to encapsulate the scale-independent assumption that a physical system's pre-existing properties can be passively observed without altering its subsequent dynamics. A necessary condition for passive realism is provided by the no-signalling in time (NSIT) condition, which states that a measurement outcome is independent of whether or not a prior measurement was performed. While closed
quantum systems generically violate NSIT due to measurement back action, we investigate whether open-system dynamics can dissipate this disturbance to the environment and restore NSIT. By modeling the time
evolution between sequential measurements as an arbitrary
quantum channel, we show that such environmental masking is possible only in two special cases. Specifically, we show that if the initial state of a two-time sequential measurement scenario is not maximally mixed, and if the channel governing the dynamics between measurements is not a discard-and-prepare channel, then there necessarily exist binary projective measurements violating NSIT, hence violating passive realism. By providing a constructive procedure for determining these violating measurements—applied here to depolarizing and dephasing channels—we demonstrate that, outside these two cases, system-environment interactions cannot erase the statistical footprint of a
quantum measurement for all measurement choices.
Read the full abstract ▾
Computation and Language
Jonathan Graehl
Privacy redaction must reliably remove personal information while preserving the relationships expressed in the text. We develop a general-purpose
multilingual named-entity tagger with distinctions fine-grained enough to support a broad range of redaction policies, and methods for cheaply learning additional distinctions. We fine-tune the
multilingual encoder XLM-R-large on text in 35
languages annotated by prompted frontier
language models, plus existing human-annotated corpora with labels mapped onto our entity types. We evaluate on human-annotated test sets in seven
languages and on LLM-annotated text in 35
languages. On the human-annotated sets, at operating points fixed on development data, our tagger reaches 88.2 redaction F1, against 69.1 for published GLiNER2 and 61.4 for Microsoft Presidio, and its architecture runs 4.9 times faster than GLiNER2's on CPU. Compared with frontier-model annotations alone, adding the mapped human-annotated corpora to training raised redaction F1 on those sets by ten points, with essentially no change on the LLM-annotated text. Open-weight LLMs fitting on one 96-GB GPU were less accurate than our tagger, whether prompted to tag text or used as encoders beneath a trained tagging layer, and roughly 30 to 1,100 times slower. We release code,
prompts and training recipes, with data-acquisition scripts and source links.
Read the full abstract ▾
Dynamical Systems
Validation of Symmetry Induced Bifurcation Points in Nonlinear Diffusion Problems
Maxime Breden, Evelyn Sander and Thomas Wanner
Nonlinear
parabolic partial differential equations frequently occur in the modeling of applied phenomena. They often exhibit complicated dynamics which is organized by their associated
equilibrium solutions. As a function of the involved model parameters, the
equilibria can be detected using a bifurcation analysis, which in turn is centered around a deeper understanding of bifurcation points. In many applied systems establishing these organizing centers rigorously is difficult to achieve using traditional mathematical techniques. In this paper, we propose a theoretical framework for computer-assisted proofs of
symmetry-induced transcritical and pitchfork bifurcations in nonlinear problems. This s accomplished by reformulating the existence question via a zero finding problem for an associated extended nonlinear system. The approach is general and allows for a wide variety of underlying
symmetries. We demonstrate how these results can be applied in the setting of
parabolic systems and higher-order equations, on both one- and two-dimensional domains.
Read the full abstract ▾
Combinatorics
A rainbow partition theorem for trees and connected maximin share allocations of chores
Marcin Anholcer, Maciej Bartkowiak, Bartłomiej Bosek et al. (+3 more)
Xiao, Qiu, and Huang (AAMAS 2023) and independently Lonc (personal communication) asked whether indivisible chores located at the vertices of a tree can always be allocated to $n$
agents in connected bundles so that the cost of every
agent is at most its connected maximin share; for goods, this is a
theorem of Bouveret, Cechlárová, Elkind, Igarashi, and Peters. We answer the question affirmatively, even for monotone costs. The answer follows from a combinatorial
theorem: if $\mathcal P_1,\ldots,\mathcal P_k$ are partitions of the vertex set of a finite tree, each into at most $k$ connected parts, then the vertex set can be split into disjoint sets $B_1,\ldots,B_k$, some possibly empty, such that each nonempty $B_i$ is connected and contained in a part of $\mathcal P_i$. Equivalently, if each of $k$ colours occurs on at most $k-1$ edges of a tree, then the vertices can be partitioned into connected sets labelled by distinct colours, none containing an edge of its own colour; in particular, one can choose for every colour a component of the forest obtained by deleting that colour so that the chosen components cover the tree. The bound is already best possible for paths, and for additive costs, the
theorem is equivalent to the fair-division statement. The proof reduces the problem to inward partitions of oriented trees, which we obtain from the colourful KKM
theorem on a simplex of edge weights, using a leaf-elimination labelling that remains compatible when weights vanish. We also give an
algorithm running in time $k^{O(k)}$ plus
polynomial time.
Read the full abstract ▾
All authors Marcin Anholcer, Maciej Bartkowiak, Bartłomiej Bosek, Jarosław Grytczuk, Zbigniew Lonc, Paweł Rzążewski