Data Structures and Algorithms
Improved Approximations for Vehicle Routing with Nonuniform Speeds
Hong Li
We study vehicle routing with vehicles of different speeds on a complete undirected graph whose vertex set consists of a depot and a set of clients, where the distances satisfy the triangle inequality. Each vehicle has a specified speed, and if the total length traveled by a vehicle of speed $s$ is $L$, its completion time is $L/s$. In the heterogeneous traveling salesman problem (HetTSP), each vehicle executes one tour starting and ending at the depot, and the tours collectively visit all clients. The objective is to minimize the maximum completion time among the vehicles. We give a $6$-approximation
algorithm for HetTSP, improving the previous $90(1+\delta)$-approximation for any fixed $\delta>0$. We also consider two versions of the heterogeneous capacitated vehicle routing problem (HetCVRP). Each client has a demand, and the vehicles have identical capacities. A vehicle may execute several tours, each starting and ending at the depot, and reload at the depot between consecutive tours; the total demand delivered on each tour cannot exceed the vehicle capacity. In the split-delivery version of HetCVRP, the demand of a client may be divided among multiple visits, possibly by different vehicles. We give a $\frac92+2\sqrt3<7.965$-approximation
algorithm for this problem. In the unsplit-delivery version of HetCVRP, the entire demand of each client must be delivered in a single visit. We give a $\frac{11}{2}+3\sqrt2<9.743$-approximation
algorithm, improving the previous $450(1+\delta)$-approximation for any fixed $\delta>0$.
Read the full abstract ▾
Robotics
MiniWAM: Learning Compact Future Targets for Efficient World-Action Modeling
Jie Chen, Ruofei Bai, Yuxin Cai et al. (+5 more)
World modeling has emerged as an effective co-training objective for
robot policies, giving rise to World Action Models (WAMs) that jointly predict actions and future states. However, most WAMs predict future states in the native representation space of pretrained visual backbones, resulting in high-dimensional targets with substantial training cost. We introduce MiniWAM, which instead predicts compact future representations learned from privileged current-future transitions. To construct these targets, we propose Predictive Representations via Inverse Spatiotemporal Modeling (PRISM), which combines inverse-dynamics supervision with feature reconstruction to emphasize control-relevant transition information while preserving useful future-state information. With the learned PRISM encoder frozen, MiniWAM is trained to jointly predict the resulting targets and
robot actions from current observations. With 65$\times$ fewer native future feature
tokens, MiniWAM consistently outperforms native future-feature prediction with both DINOv3 and WAN2.1 VAE features, while achieving up to an 8$\times$ speedup in world-action training. At 0.25B parameters, MiniWAM is already competitive with substantially larger WAMs on LIBERO, LIBERO-Plus, and RoboTwin 2.0 simulation
benchmarks. Representation analyses further show that PRISM contributes behavioral structure beyond feature reconstruction alone. These results demonstrate that effective world-action modeling does not require predicting native visual futures, and that compact predictive representations provide a strong and substantially more efficient target for policy learning. The project page is available at: https://j1dan.github.io/MiniWAM.
Read the full abstract ▾
All authors Jie Chen, Ruofei Bai, Yuxin Cai, Yifeng Zhang, Chengyang He, Jun Li, Wei-Yun Yau, Guillaume Sartoretti
Machine Learning
DataSense-Bench: The First Step Toward an AI Scientist
Yudi Zhang, Mingyu Cao, Lu Yin et al. (+2 more)
As claims about recursive self-improvement (RSI) and artificial general intelligence (AGI) proliferate, we ask a simple question: do frontier AI models have a sense of data, i.e., can they reliably select the right data for training? We introduce DataSense-Bench to study this capability through the fundamental problem of data selection and performance forecasting in machine learning. We ask AI
agents to select and rank candidate training subsets that can be used to fine-tune a small LLM model.
Agents are allowed to inspect the data, write and execute analysis code, and run model forward passes, but can not train the model or access the actual evaluation tasks. We then fine-tune the base model on each selected subset and evaluate its post-training performance under a standardized protocol. We instantiate the
benchmark in terminal problem solving and tool use, selecting trajectories from OpenThoughts-
Agent and EnvScaler and evaluating on TBLite and BFCL, respectively. We then evaluate the
agents along two complementary dimensions: the post-training performance of the top-ranked subset, reflecting the ability to identify high-value training data, and ranking accuracy, reflecting the ability to predict the relative performance of the selected subsets. In our experiments, selection gains over random selection are limited;
agents do not reliably rank their selected groups, and ranking ability does not hold consistently across tasks: Astra identifies the best group in all three tool-use runs but in only one of three terminal runs. Analysis of execution traces on both tasks shows that
agents often use similar data signals while interpreting their training value differently.
Read the full abstract ▾
All authors Yudi Zhang, Mingyu Cao, Lu Yin, Mykola Pechenizkiy, Shiwei Liu
Machine Learning
Just Weather Scoring: Efficient End-to-end Nowcasting with Distributional Diffusion
Jannik Wiese, Johannes Schusterbauer, Tommaso Martorella et al. (+1 more)
Generative
diffusion models are well-suited for probabilistic precipitation nowcasting, but existing approaches often rely on separately trained compression or deterministic forecasting components and remain costly at inference due to iterative denoising. We introduce Just Weather Scoring (JWS), a single-stage, end-to-end
diffusion model which addresses both issues by forecasting directly in radar space and enabling few-step generation. Radar-space modeling greatly simplifies training and inference and eliminates uncertainty arising from lossy compression. JWS combines Masked Asynchronous
Diffusion, a timestep-sampling scheme that preserves clean context while adapting
diffusion training to high-dimensional spatio-temporal data, with a simple scoring-rule objective that aligns training with probabilistic forecasting and unlocks few-step generation. On the SEVIR and MeteoNet
benchmarks, JWS achieves state-of-the-art probabilistic forecasting performance at reduced training and inference cost. Even our smallest model remains competitive using substantially fewer parameters and more than 17x faster inference.
Read the full abstract ▾
All authors Jannik Wiese, Johannes Schusterbauer, Tommaso Martorella, Björn Ommer
Robotics
RESETTLE: Robotic Recovery through Disagreement-Triggered Retrieval and Efficient Corrective Control
Yuxin Chen, Senqiao Yang, Zixuan Wang et al. (+6 more)
Reliable robotic manipulation requires timely intervention to correct emerging deviations and restore progress after execution errors. However, recovery methods based on repeated
vision-
language reasoning or iterative online optimization can incur substantial latency, delaying intervention. To address these challenges, we introduce RESETTLE(Robotic rEcovery through diSagrEement-Triggered
reTrievaL and Efficient Corrective Control), a model-agnostic framework that provides computationally efficient recovery at the action-execution interface of frozen
robot policies. RESETTLE triggers recovery when two action proposals independently sampled under identical conditioning persistently disagree. It retrieves a same-task demonstration reference using an adapted V-JEPA encoder and combines a state-servo prior with a guarded visual residual to execute one corrective action without online trajectory optimization or additional
vision-
language reasoning, then returns control to the base policy. Across six base policies in simulation, RESETTLE achieves up to 8.70%, 6.28%, and 6.83% absolute success-rate gains on LIBERO-Plus, Meta-World, and RoboCasa Tabletop, respectively, with further improvements on four real-world tasks using two policies. In QwenPI-based comparisons, its monitoring-and-recovery computation latency is 74.04%–93.57% lower than VoLoAgent's monitoring-and-planning latency for grasp and place tool calls. It also raises Harness VLA's LIBERO-Pro Swap success from 42% to 50%, demonstrating compatibility with high-level agentic planning. Code available at: https://github.com/JIA-Lab-research/RESETTLE
Read the full abstract ▾
All authors Yuxin Chen, Senqiao Yang, Zixuan Wang, Jinhui Ye, Changsheng Lu, Pengguang Chen, Shu Liu, Zhuotao Tian, Jiaya Jia
Machine Learning
A Closer Look at Agentic BBO: Benchmarking LLM Agents for Black-Box Optimization
Ming Chen, Rong-Xi Tan, Ke Xue et al. (+8 more)
Black-box optimization (BBO) arises in many scientific and engineering problems where objective evaluations are expensive and limited. Recent large
language model (LLM)
agents offer a new way to approach BBO by combining task semantics, computation, optimization tools, and feedback-driven decision making, showing great potential due to the integration with mathematically rigorous tools. However, existing agentic BBO studies use different task domains and system configurations, making their results difficult to compare and the effects of individual design choices hard to isolate. We therefore introduce AgenticBBO-Bench, a cross-domain
benchmark for agentic BBO spanning synthetic functions, hyperparameter optimization, database tuning, chip design, and molecular design under a unified finite-budget evaluation protocol. In our experiments, agentic BBO achieves higher family-averaged scores than direct LLM-based methods in all five domains and outperforms the best numerical optimizers in four. We further study three factors shaping
agent performance: optimization tools, task information and prior knowledge, and the role of the LLM during search. Our results show that additional numerical tools do not consistently improve performance, task semantics are broadly useful while more specific priors are less reliable, and numerical optimizers can effectively absorb gains from search trajectories established by the
agent. Finally, we introduce a five-task frontier challenge within AgenticBBO-Bench and evaluate seven LLMs under the Codex
agent harness, where GPT-6 Astra and DeepSeek-V4.1-Flash lie on the Pareto frontier of performance and cost among the evaluated models. Our code is available at https://github.com/lamda-bbo/agentic-bbo.
Read the full abstract ▾
All authors Ming Chen, Rong-Xi Tan, Ke Xue, Yu-Jie Zhou, Taiye Lu, Zhi-Xuan Gao, Peng Xie, Zijun Shen, Chen Lu, Haopu Shang, Chao Qian
Computer Vision
AI-Based On-Board Maritime Object Detection for Earth Observation Payload Data Reduction on Versal Embedded Hardware
Thomas Goudemant, Aurélien Bobey, Omar Hlimi et al. (+1 more)
Very-high-resolution Earth-observation satellites acquire more data than they can store and downlink, while in maritime surveillance the vessels cover a tiny fraction of each scene. We study onboard vessel detection as a way to select what is downlinked, which reduces the data according to its content rather than coding every pixel; it is complementary to conventional onboard compression. The work follows three axes. (i) Data and
algorithm: a controlled
dataset is generated from 68 annotated Maxar scenes with 43 vessel classes, and a YOLOX-S detector is trained on it. (ii) Embedded deployment: the detector is quantized and deployed on the DPU of a Versal VC1902, with a limited loss of detection quality and a processing time of a few seconds per scene. (iii) Data reduction: we propose several downlink modes, from metadata only (box, class and score of each detection) to image crops around vessels, tiles holding detections, or the whole scene with a degraded background, and estimate from the measured detection errors the trade-off each offers between the vessels kept and the volume downlinked. On our dense harbor and coastal scenes, tiles keep 98% of the vessels with 29% of the scene volume, and crops 83% with 3%.
Read the full abstract ▾
All authors Thomas Goudemant, Aurélien Bobey, Omar Hlimi, Marjorie Bellizzi
Information Theory · Cross-listed from eess.SP
Localization in Multi-Panel Massive MIMO With Clock Asynchronism: A Unified Approach
Yuhao Zhang, Guangjin Pan, Musa Furkan Keskin et al. (+3 more)
In this work, we propose a unified localization framework, termed UNILocMP, that combines model-based geometry and channel charting (CC) for multi-panel massive multiple-input-multiple-output (
MIMO) under clock asynchronism. Owing to the multi-panel architecture, users are classified into multi-line-of-sight (LoS) users, which maintain LoS links with at least two panels, and single/non-LoS users, which maintain a LoS link with only one panel or with none of the panels. For multi-LoS users, a joint position and clock bias estimation is developed; while for single/non-LoS users, an unsupervised CC model is trained with a two-stage data augmentation strategy, where a clock-bias aware dissimilarity metric is introduced. It is numerically validated that the proposed UNILocMP outperforms model-based and CC-based baselines and achieves acceptable performance compared with fully-supervised fingerprinting. Moreover, for a fixed total number of
antennas, the multi-panel architecture significantly improves localization accuracy and robustness compared with a single-panel base station (BS) deployment, particularly in the presence of clock asynchronism.
Read the full abstract ▾
All authors Yuhao Zhang, Guangjin Pan, Musa Furkan Keskin, Carmen D'Andrea, Stefano Buzzi, Henk Wymeersch
Artificial Intelligence
Recursive Self-Improvement through Multi-Agent Self-Supervision
Hyunin Lee, Jinglue Xu, Jeffrey Seely et al. (+4 more)
Recursive self-improvement (RSI) of a model on non-verifiable tasks, such as open-ended research, faces a supervision bottleneck when its outputs exceed what even human experts can reliably assess, leaving the model itself (optimizee) as the best available optimizer and evaluator. However, a single model instance struggles to critique and improve its own complex
reasoning under this homogeneous loop. To address this, we propose Multi-
Agent Self-Supervision (MASS), an RSI method that alternates between evolutionary workflow optimization and supervised fine-tuning on self-generated trajectories. Guided by early findings that multi-
agent topologies excel at complex
reasoning, MASS
prompts a single base model to iteratively propose, execute, and self-evaluate multi-
agent workflows. Through an evolutionary search constrained by structural guardrails, the model optimizes these computational-graph-like orchestrations, discovering the most effective distinct roles and information routing for a given task. Over two MASS cycles with Qwen3.6-27B, the model achieves 1.2-1.6x higher performance per output
tokens on four open-ended public
benchmarks. Because the improved model subsequently acts as a better optimizer and evaluator, this alternating framework enables a continuous, recursive bootstrapping of the model's capabilities. Moreover, multi-
agent traces are also more training-efficient: a student trained on them outperforms a single-
agent student trained on 1.4x more training
tokens. These findings suggest that jointly learning orchestration and bounded subagent execution from multi-
agent trajectories can provide an effective signal for RSI.
Read the full abstract ▾
All authors Hyunin Lee, Jinglue Xu, Jeffrey Seely, Donghyun Lee, Somayeh Sojoudi, Matei Zaharia, Yujin Tang
Robotics
Unifying Policy Learning and State Prediction through Spatial Language Modeling
Minye Wu, Zehao Wang and Tinne Tuytelaars
Learning how actions change scene geometry can provide complementary supervision for goal-directed manipulation. We introduce Spatial
Language Modeling, which represents scene contours, goals, action targets, and future states with a shared vocabulary of discrete coordinates and semantic
tokens. A task-specific grammar organizes these elements into spatial sequences, allowing one autoregressive
Transformer to learn action generation and action-conditioned state prediction through a common next-
token objective. We train the model from scratch using random-play transition pretraining followed by joint action and state training on expert demonstrations. During pretraining, recorded action coordinates condition subsequent state predictions and are excluded from the prediction loss. During control, the model decodes only executable action targets and updates its history with newly observed states. We evaluate the approach on Push-T in simulation and on a real
robot. The model achieves competitive simulation performance and higher task success and target coverage than the evaluated real-
robot policy baselines. Training ablations show improved control with joint action and state sequences, with further gains from random-play pretraining. Given supplied action trajectories, the same model also predicts successive scene states, capturing the
geometric effects of pushing.
Read the full abstract ▾
Artificial Intelligence
Learning to Plan by Looking Back: Hindsight Hierarchies for Training Reasoning Models
Lars Simon, Holger Eble and Manuel Radons
We introduce a self-improvement loop for
reasoning models based on the following observation: Even when the difficulty of a problem exceeds the model's current solving abilities, an additionally supplied solution might enable the model to extract useful solution ideas in hindsight. We operationalize this by jointly training the same model to exhibit the following three capabilities: predicting solution ideas from problems alone, reverse-engineering ideas from problems and known solutions, and solving problems using provided ideas. The loop alternates between reverse engineering such ideas from problems with supplied solutions and using these ideas as additional supervision for joint training of all three capabilities. We give a formal specification of our method and a concrete instantiation for interactive
theorem proving in the Lean
theorem prover; empirical evaluation remains future work.
Read the full abstract ▾
Machine Learning
Is Real-World Training Data Necessary for Generalist Graph Anomaly Detection?
Yujing Liu, Yixin Liu, Yue Tan et al. (+4 more)
Generalist graph anomaly detection (GAD) aims to build a foundation model that detects anomalies on arbitrary unseen graphs without retraining or fine-tuning. Sufficient data are essential for foundation model training, yet generalist GAD still faces a data shortage, as real-world anomalous graphs are scarce and costly to collect and annotate. To fill this gap, we propose AG-FORGE, an Anomalous Graph generation Forge for automatic synthesis of anomalous graphs, exploring the feasibility of synthetic data-driven training for generalist GAD. Empirically, we find that synthetic data can achieve performance comparable to real-world training, but fail to push the performance boundary further due to the limited capacity of existing methods. To further unlock model capacity as training data scale up, we develop TS-GGAD, a
Topology-Semantic coordinated Generalist GAD that captures complementary topological and semantic anomaly evidence, together with a curriculum learning strategy tailored to large-scale synthetic training. Extensive experiments on 14 real-world
datasets demonstrate that TS-GGAD, trained on data generated by AG-FORGE, significantly outperforms state-of-the-art methods.
Read the full abstract ▾
All authors Yujing Liu, Yixin Liu, Yue Tan, Xiaofeng Cao, Alan Wee-Chung Liew, Heng Tao Shen, Shirui Pan
Operating Systems
Mole: Tier-Specific Hotness Profiling Driven Memory Tiering for Multi-Tiered Memory Systems
Mingyang Liu, Congming Gao, Xufeng Yang et al. (+4 more)
Multi-tiered
memory systems combine fast, small upper tiers with slow, large lower tiers to improve performance and cost efficiency. Existing designs, such as AutoTiering and MTM, rely on greedy promotion and stepwise demotion, which can intensify contention for limited capacity in faster tiers. We observe that promotions directly improve performance, whereas demotions primarily reclaim space. Based on this asymmetry, we propose Mole, a
memory tiering system that separates promotion and demotion destinations. Mole employs staging demotion to move cold pages directly to the lowest tier, bypassing intermediate tiers, and targeted promotion to place hot pages in tiers that match their current hotness. The lowest tier serves as a staging area from which pages can be promoted when they become hot again. This separation preserves intermediate-tier capacity for promotions, reducing tier contention and migration failures. However, staging demotion requires timely identification of pages that become hot after demotion. To meet this requirement with low profiling overhead, Mole uses tier-specific profiling: it profiles the lowest tier at high frequency to detect reactivated hot pages, while profiling upper tiers at lower frequency to identify cold pages. Experimental results show that Mole reduces migration failures and improves performance under dynamic access patterns.
Read the full abstract ▾
All authors Mingyang Liu, Congming Gao, Xufeng Yang, Fang Wu, Youmin Chen, Renhui Chen, Jiwu Shu
Machine Learning
Scalable Hierarchical Graph Generation via Soft Community Structure
Ahmet Tüzen, Helge Langseth and Kjetil Nørvåg
Generating large attributed graphs requires reproducing the
topology, generating attributes jointly with the structure, and remaining scalable. Many real-world graphs exist as a single large graph, so a generative model has to generalize from the one graph it is fit on, without independent samples. We present Schema, which recursively decomposes a reference graph into a hierarchy of soft communities, assigning each node a membership distribution. Generation is then split into three stages, each trained independently: (1) synthesizing node attributes conditioned on soft memberships, (2) generating intra-community edges from local structural context, and (3) modeling inter-community connections over bridge nodes whose membership mass is distributed across several communities. No stage forms the full adjacency
matrix, and each stage operates on a subgraph bounded by the community size. We also introduce an evaluation protocol that covers structural fidelity, memorization, downstream utility, and scalability. On four real-world attributed graphs, Schema recovers the balance between local and long-range structure more closely than any other model that generates attributes, while reproducing only a small fraction of the reference edges. It retains the downstream accuracy of the reference graph without raising it artificially above that level. Baselines that match its structural fidelity memorize the reference, while those with higher downstream accuracy either exceed the reference accuracy or fail to complete on the larger graphs. We measure scalability on six additional graphs with up to 10 million nodes.
Read the full abstract ▾
Machine Learning
When KL Regularization Misfires in Group Policy Optimization
Fei Ding
Why does removing reference-policy KL regularization sometimes improve group policy optimization? This motivates studying how reference-policy information should enter group-relative updates. We analyze seven potential failure modes in the interactions between KL and rewards: residual KL updates after reward clipping, after gradient cancellation, and in groups with identical rewards; KL growth with response length and an imbalance in its relative contribution; KL concentration on a small number of
tokens; and sampling noise when k1 is incorporated into rewards. We propose Zero-Sum Calibrated Policy Optimization (ZCPO), which uses relative drift measured by conditional KL to calibrate within-group reward coefficients and integrates them into the base surrogate. Mathematical
reasoning experiments and ablations support this design's effectiveness in our settings.
Read the full abstract ▾
Robotics
Instance-anchored interaction evidence: Grounding robot plans in human pointing and handling
Xinliang Xiao, Bowen Yang, Wenjing Zhang et al. (+2 more)
A
robot that assists people must often act on what a person has shown rather than said: which of several identical cartons was pointed at, or which box was handled. The plan is executed from the final scene, whereas the evidence occurs earlier, possibly on objects that have since moved. We propose instance-anchored interaction evidence (IAE), which registers every object of the final scene to its public identifier, keeps each identity through the
video by backward mask propagation, and describes every frame by the geometry between hands, forearms and these instances. An evidence network trained only from task outcomes scores the instances. For pointing tasks, a grammar-constrained dynamic program trained with a structured loss decodes object-destination programs; symbolic programs handle reference disambiguation and, without learning, episodic tasks. On 1,255 WatchAct
benchmark requests, scored by symbolic execution, IAE reaches 64.2% plan success on implicit-intent tasks against 27.5% for a 32B
vision-
language model (strict success 49.7% against 15.4%), and 46.4% against 27.0% on restoration, reversal and imitation without task-specific training. Controls with the same perception overlays, the same 32 frames, forward tracking, or a relation model trained on the same labels do not explain the gain. Given IAE's evidence as text with its meaning explained, the same
language model reaches 57.4%: most of the gain comes from the instance-anchored evidence, and the explicit programs add 6.8 points at a fraction of the cost. Pointing remains the hardest case, with 16.9% strict success. The code is available at https://github.com/WeiZhou96/iae-watchact.
Read the full abstract ▾
All authors Xinliang Xiao, Bowen Yang, Wenjing Zhang, Li Yang, Wei Zhou
Computer Vision
Connected Self Forcing: Beyond Local Learning in Video Autoregression
Dongbin Zhang, Chaoda Zheng, Kangjie Chen et al. (+13 more)
To stream long
videos while maintaining visual quality and temporal consistency, Self Forcing mitigates exposure bias through self-rollout training on self-generated histories with key-value (KV) caching. To keep
memory manageable, it detaches historical caches, preserving forward dependencies between chunks but severing the backward gradient paths. We introduce Connected Self Forcing, a training framework that reconnects gradient paths across autoregressive chunks, allowing feedback from later predictions to guide how earlier context is generated. These connections go beyond historical KV-writing: gradients pass through generated latents into the computations that produced them, linking the generation of earlier context to its use in later predictions. To make this connected training
memory-efficient, we develop shortcut gradient replay, which recovers cross-chunk gradients without retaining the full rollout computation graph. Integrated with distribution matching
distillation, Connected Self Forcing trains historical chunks according to both their direct supervision and their contribution to subsequent generation. Experiments on autoregressive
video generation show improvements in long-horizon visual quality and temporal consistency, without changing the inference procedure.
Read the full abstract ▾
All authors Dongbin Zhang, Chaoda Zheng, Kangjie Chen, Xiangyu Li, Shijia Chen, Jinhao Deng, Yuqi Zhang, Guangfeng Jiang, Hongbin Lin, Choo Sin Wai, Minqi Wang, Puyi Wang, Jingye Zhang, Yu Zhang, Xianming Liu, Boyang Wang
Systems and Control
Stochastic Distribution Network Reconfiguration under Load Uncertainty
Celso Cavellucci and Fábio Luiz Usberti
This paper investigates distribution network reconfiguration under demand uncertainty using a two-stage
stochastic formulation. The network
topology is selected in the first stage, whereas the electrical variables are determined separately for each scenario. The deterministic equivalent is formulated as a MISOCP problem that accounts for power balances, losses, voltage constraints, and radiality. Experiments on 12 networks, ranging from 17 to 10,561 nodes, consider three demand scenarios. Reconfiguration reduced expected losses in all instances, with reductions ranging from 3.23% to 65.23% and averaging 31.76%, while also reducing the number of voltage violations in the ex post evaluation by approximately 55%. Eleven of the 12 problems were solved to optimality within the prescribed time limit. However, the additional benefit of
stochastic modeling was negligible because of the homothetic structure of the scenarios considered, which preserves the spatial distribution of loads. The results show that spatially heterogeneous scenarios are important for the representation of uncertainty to meaningfully affect reconfiguration decisions.
Read the full abstract ▾
Machine Learning
Toward Optimal Regret in Adversarial MDPs with Stochastic Hard Constraints
Qian Zuo and Francesco Emanuele Stradi
We study episodic constrained Markov decision processes with
adversarial losses under
stochastic hard constraints. Specifically, starting from a known strictly feasible policy with margin $d$, we seek to obtain optimal regret while satisfying the expected cost constraints in every episode. In this setting, Stradi et al. (2025) show that a carefully designed mixing rule attains regret of order $\widetilde{\mathcal{O}}(\sqrt{T}/\min\{d,d^2\})$. Interestingly, they also provide a lower bound of order $\Omega(\sqrt{T}/\rho)$ for the same setting, where $\rho$ is the Slater margin of the offline problem and can be much larger than $d$. In this work, we build on their approach to obtain optimal regret dependence on these margins. Specifically, we propose MA-OPS, an
algorithm that combines an optimistic search for the Slater margin with a pessimistic evaluation of the selected policies to safely learn a policy with a large feasibility margin. This policy is then used to minimize regret while satisfying the constraints at every episode. In particular, we show that MA-OPS attains regret $\widetilde{\mathcal{O}}(\sqrt{T}/\rho + 1/(d\rho))$. Finally, we provide a matching lower bound, showing that the dependence on $T$, $d$, $\rho$ in the regret bound is optimal up to logarithmic factors.
Read the full abstract ▾
Numerical Analysis
Variational Derivation of the Weak Form for Viscous Flows Involving Free Surfaces with Implications for Geodynamic Simulations
Karim Norouzi Moghanjoghi, Javier Garcia Pintado, Daniel R. Shapero et al. (+1 more)
Geodynamic simulations involve solving the Stokes equations, where boundary conditions play a crucial role in governing internal dynamics and surface topography
evolution. A common approach is to model the interface in contact with air as a free surface. This interface is stress-free and deforms in response to internal force balances, which is essential for understanding processes ranging from rifted margins to subduction. However, a free surface can introduce instabilities and oscillations in the numerical solution, commonly known as the "drunken sailor" instability, forcing the use of small time steps. Geodynamic models therefore often employ stabilization terms based on the gravitational effect of free-surface motion. To date, these terms have been derived primarily on a heuristic basis. Here, we apply variational calculus to the energy-minimization formulation of incompressible Stokes flow. Using the Gateaux derivative, we derive the complete weak form for a deforming free-surface domain and show that the stabilization terms arise naturally from the free-surface assumption. We consider the linear-viscosity case and identify two boundary-integral contributions: one associated with internal viscous deformation and the other with gravitational body force. Two
benchmark cases demonstrate the consistency of the derived formulation with previously proposed stabilization approaches. Our results provide a mathematical foundation for the commonly used
gravity stabilization and identify an additional viscous boundary contribution.
Read the full abstract ▾
All authors Karim Norouzi Moghanjoghi, Javier Garcia Pintado, Daniel R. Shapero, Marta Perez-Gussinye
Machine Learning
Bayesian Optimisation under State-Preservation Constraints
Gabriel Diaz-Aylwin, Vignesh Gopakumar, Omkar Myatra et al. (+4 more)
In many engineering design problems, the objective and constraints depend on the state: the solution of a PDE determined by the design parameters. We consider improving a design while holding selected state observables near trusted values, which we call state preservation constraints. Constrained
Bayesian optimisation handles these with a learnt feasibility model, but struggles with this problem's highly anisotropic feasible set. Our central idea is to pre-compute the set of controls whose linearised constraint response stays within tolerance, thereby pulling back the state-space constraint into design space. This linearisation defines an ellipsoid from which we can efficiently draw a large number of well-spread candidates. The underlying linear response map is refined online, and the ellipsoid is rebuilt accordingly. We demonstrate the method end-to-end on our key application - Tokamak divertor optimisation under
plasma-boundary preservation.
Read the full abstract ▾
All authors Gabriel Diaz-Aylwin, Vignesh Gopakumar, Omkar Myatra, David Moulton, Lorenzo Zanisi, David S. Leslie, Henry B. Moss
Software Engineering
Reliability Characterization for N-version Object Detection
Shunsuke Nagao and Fumio Machida
N-version object detection (OD) is an approach to diversifying detection results using multiple models or input frames and reducing detection errors by aggregating individual results. Diversity and consistency across multiple detection results are critical information for characterizing the reliability of possible configurations of N-version OD systems. However, existing performance metrics such as mAP and Accuracy fail to capture these factors, as they are defined solely on the final outcome after aggregation. To overcome this limitation, we propose two reliability metrics particularly defined for N-version OD, namely coverage of errors in OD (Cov_OD) and certainty of accurate prediction in OD (Cer_OD), which can be computed from individual detection results without relying on voting strategies. We empirically demonstrate the unique features of the proposed metrics through a case study of N-version OD for a vehicle in an autonomous-driving simulator. We show that the proposed metrics can be used for 1) selecting version combinations with complementary error characteristics, 2) choosing an effective voting strategy based on diversity and consistency profiles, and 3) guiding incremental construction of N-version OD systems via pairwise two-version analysis. These results highlight the importance of the metrics that can guide the design of reliable N-version OD applications.
Read the full abstract ▾
Machine Learning
Large-Scale Benchmarking of Quantum Neural Network Configurations for Financial Time Series Forecasting
Jack Waller, Xing Liang, Dimitrios Makris et al. (+1 more)
Quantum machine learning, and
quantum neural networks (QNNs) in particular, are advancing fields with growing potential. Although systematic comparisons of QNN configurations have been explored primarily for classification tasks, comparatively little
attention has been given to regression problems, particularly financial time series forecasting. This study presents a large-scale systematic comparative evaluation of QNN component configurations for financial time series forecasting, using the GBP/USD spot exchange rate as a case study. A grid search across encoding methods, ansatz designs,
qubit counts, layer depths, and cost functions yields 1,368 distinct model configurations, each evaluated in terms of prediction accuracy, computational cost, and convergence behaviour. The results reveal unique insights into how the choice of methods influences performance, such as that gate selection and arrangement are more critical to model success than raw parameter count, and that
entanglement is a system-level property of the full
circuit rather than solely at the ansatz level. The best-performing QNN configuration achieves an $R^2$ score of 0.985, outperforming a classical BiLSTM baseline. Additionally, the impact of real
quantum hardware noise is assessed through execution on the IQM Emerald device, revealing that gate errors and decoherence represent a significant barrier to practical deployment, with gate selection and
circuit depth identified as key determinants of hardware noise resilience. Overall, the findings provide practical architectural guidance for QNN design and establish a baseline characterisation of QNN noise sensitivity on near-term
quantum devices.
Read the full abstract ▾
All authors Jack Waller, Xing Liang, Dimitrios Makris, Rajagopal Nilavalan
Computer Vision
Healthy Counterfactual Generation via Diffusion Inpainting for Mammography Classification
Inês Cruchinho Garcia, Mariana Mourão, Francisco Maria Calisto et al. (+2 more)
False negatives remain a critical limitation of computer-aided diagnosis (CAD) systems for breast cancer screening due to delayed detection and treatment. To address this issue, we propose a counterfactual data augmentation strategy that generates healthy mammograms by "erasing" lesions from anomalous images, thereby enriching the training distribution. We train a Denoising
Diffusion Probabilistic Model on BI-RADS 1 (healthy) mammograms and use a RePaint-based sampling strategy to inpaint realistic normal tissue within annotated lesion bounding boxes. The resulting healthy counterfactuals replace annotated lesion regions with realistic healthy tissue while preserving patient-specific anatomical structure, as supported by similarity metrics between real and generated images. Image realism was further assessed by radiologists and found to be consistent with the original
dataset quality. We evaluate counterfactual augmentation across four representative classifier architectures: a convolutional neural network (ConvNeXt), a
vision transformer (ViT), a
vision-
language model pre-trained on mammogram-report pairs (Mammo-CLIP) and a multi-scale
attention-based multiple-instance learning framework (FPN-MIL). Experiments conducted on the VinDr-Mammo
dataset show improvements in sensitivity across all architectures, particularly at 80% fixed specificity, contributing towards more reliable CAD systems for breast cancer. Code is available at: https://github.com/ines03garcia/
diffusion-based-counterfactual-generation.
Read the full abstract ▾
All authors Inês Cruchinho Garcia, Mariana Mourão, Francisco Maria Calisto, Carlos Santiago, Jacinto Nascimento
Computation and Language
Language-Specific Effects of Tokenizer Choice in Multilingual Language Models
Clara Meister, Gül Sena Alt{\i}ntaş and Antoine Bosselut
Tokenizer choice affects multilingual
language modeling, but vocabulary capacity is finite and vocabulary size is often constrained: improving representation for some
languages often comes at the expense of others. We therefore ask whether tokenizer choice matters equally across
languages, a question that the current literature leave unanswered. To this end, we train 123
language models spanning 54 tokenizers. In the main comparison, architecture, training corpus, training-
token budget, and optimization are held fixed, so the models differ only in their tokenizer. We find that tokenizer choice matters more for
languages with less
language-model training data: across the 54 tokenizers, the standard deviation of a
language's bits-per-byte (BPB) increases as its model training-data share decreases (Spearman rho = -0.52 over the 31 trained
languages and -0.69 over the 28 written with word boundaries). Leaving a
language out of tokenizer training raises its BPB in every
language we study, and the penalty tends to be larger for
languages with less
language-model training data. Giving lower-resource
languages a larger share of tokenizer-training data, however, does not unconditionally help those
languages: both equal weighting and an allocation inverting the shares with respect to the
language model training data increase their BPB, particularly when
language-model training repeats data. Finally, which intrinsic tokenizer properties are associated with better BPB differs across
languages, providing further evidence that what makes a good tokenizer depends on the
language. We find that the metrics quantifying these properties can be successfully used to predict downstream models' pairwise BPB rankings, suggesting a practical strategy for screening tokenizer candidates before training
language models.
Read the full abstract ▾
Artificial Intelligence
Instruction-Conditioned Electromagnetic Spectrum Understanding via Budget-Adaptive Signal Tokenization
Lei Zhai, Zhihao Chang, Shuyuan Yang et al. (+1 more)
Electromagnetic spectrum monitoring increasingly requires flexible analysis beyond task-specific recognition and detection.
Multimodal large
language models offer a unified interface, but extending
vision-
language models (VLMs) to raw I/Q signals requires tokenization that balances fidelity against a strict budget. For signals, dense encoding causes
token costs to grow with observation length, whereas fixed-resolution compression may discard short-duration or localized signal evidence. Thus, we propose BATok, a budget-adaptive signal tokenizer that adjusts
token capacity to the input length while allocating that capacity according to the signal content. BATok constructs candidate representations from signal-derived features using lightweight multi-resolution branches, then combines a local energy prior with learnable queries to resample these representations into compact signal
tokens. The number of
tokens adapts to the input length while remaining strictly bounded. The resulting
tokens are projected into the
language embedding space of VLMs. We further introduce EMSpec-Instruct, a
multimodal instruction
dataset aligning raw I/Q signals, waterfall images, and
language supervision for modulation recognition, structured detection, and
language-conditioned signal grounding. Experiments show that BATok learns effective signal representations and achieves competitive performance across all tasks.
Read the full abstract ▾
All authors Lei Zhai, Zhihao Chang, Shuyuan Yang, Zhixi Feng
Robotics
Leveraging Human-In-The-Loop Demonstrations in Reinforcement Learning for Digital Twin-Driven Robot Flexibility
Yuzhu Sun, Mien Van, Nguyen Minh Nhat et al. (+2 more)
Growing automation makes collaborative
robots work in more variable environments, increasing the need for adaptation. We propose a human-in-the-loop online training framework combining a digital twin (DT), reinforcement learning (RL), and human demonstrations. Unlike DTs used mainly to generate synthetic data before task execution, our DT is synchronized with the physical system in real time through camera feeds, allowing the virtual
robot to update its observations and policy from real-world feedback. A dual actor framework integrates imitation learning (IL) without adding a direct imitation loss to the RL actor, so demonstrations can guide adaptation instead of manual reprogramming. The proposed framework is demonstrated on the Ufactory Xarm5 collaborative
robot, where the
robot's end-effector aims to reach the target position while avoiding obstacles. The experiments show that the framework can resume training after a change in the physical workspace and that, with a fixed set of non-optimal demonstrations, the dual actor framework achieves a much higher final success rate than two methods that add an imitation loss to the actor. The same pattern holds with real human demonstrations collected in virtual reality (VR): with demonstrations that never reach the goal, the dual actor framework reached 83-100% mean deterministic evaluation success, against 0-17% for the two imitation-loss methods.
Read the full abstract ▾
All authors Yuzhu Sun, Mien Van, Nguyen Minh Nhat, Stephen McIlvanna, Sean McLoone
Networking
Toward Intelligent Networks via AI-Aware GPU-Native Packet Processing
Seyed Mohammad Mehdi Mirnajafizadeh, Yiwen Hu and Rhongho Jang
The integration of inline Artificial Intelligence (AI) models into critical network infrastructure is fundamentally bottlenecked by the high latency and synchronization overhead of CPU-mediated packet processing. While legacy GPU offload and recent CPU-bypass frameworks attempt to bridge this gap, they remain trapped in proprietary ecosystems or still rely on the host CPU and coarse-grained batching to coordinate stateful telemetry and AI pipeline execution. In this paper, we propose AGP, a novel framework that promotes the GPU from a passive accelerator to a primary data-path controller, answering what changes architecturally when the GPU autonomously owns the complete packet-to-inference pipeline. By enabling GPU-native packet processing, contention-free in-GPU stateful aggregation, and a persistent mega-kernel for continuous AI inference, AGP removes the CPU from the critical path. Our evaluation demonstrates that native in-GPU packet processing sustains line-rate throughput while being over 6.1x more power-efficient. Furthermore, our integrated Intrusion Detection System (IDS) stress test eliminates the legacy CPU-GPU synchronization tax, reducing end-to-end latency by 7.9x (up to 35x p99) and accelerating whole-system inference throughput by 9.3x using only ~2% of the GPU's thread capacity.
Read the full abstract ▾
Cryptography and Security
Poster: A Preliminary Study of LLM Distillation Inference
Edward Chen and Yuntao Du
Unauthorized model
distillation, in which a model is trained on the outputs of a proprietary large
language model (LLM), is a growing threat to model providers. We study
distillation inference: determining whether a suspect model was distilled from another model or trained independently. We formulate this problem as a hypothesis test and estimate the behavior expected under each hypothesis by training shadow models: distilled shadow models learn from the teacher's
reasoning traces, whereas independent shadow models learn only from reference answers. The auditor measures how closely each model predicts the teacher's
reasoning outputs and then uses the shadow models to convert the suspect's score into a calibrated p-value. In a preliminary study using Qwen2.5-7B as the teacher and Llama-3.2-3B for the suspects, our test achieves a true positive rate of 1.0 at a significance level of 0.02. These results demonstrate the feasibility of using
distillation inference to detect
distillation attacks.
Read the full abstract ▾
Artificial Intelligence
Q-Shaped Options for Hierarchical Reinforcement Learning
Clarisse Wibault, Antoine Gorceix, Antonio Léon Villares et al. (+5 more)
Learning to tackle long-horizon, goal-conditioned tasks requires an
agent to reason over extended timescales and act across a broad range of states. In principle, Hierarchical Reinforcement Learning (HRL) addresses both challenges through the interaction between action (temporal) and state (spatial) abstraction. First, using an action abstraction to represent temporally extended behaviour as options reduces the effective decision horizon. Second, enabling different state abstractions at each level of the decision process permits greater data aggregation for learning. However, realising these two benefits of a hierarchical policy depends on learning an appropriate action abstraction. Current HRL
algorithms fail in one of two ways. Some discard distinctions between options needed for optimal control, undermining hierarchy altogether. Others retain unnecessary distinctions, preserving horizon reduction, but forfeiting coarser state abstraction. In this work, we characterise three desiderata for an action abstraction. We introduce Q-Shaped Options (QSO) to address all three. QSO builds on an architecture with distinct state-value functions, Q functions and policies at each level of the hierarchy. It learns the action abstraction between consecutive levels as a shared encoder shaped by their respective Q functions. The low-level Q function uses the option as a goal, encouraging the abstraction to retain distinctions necessary for optimal control. The high-level Q function uses it as an action, encouraging unnecessary distinctions to be discarded. Across offline goal-conditioned locomotion and manipulation environments, QSO learns semantically meaningful option spaces and outperforms baselines, achieving non-zero performance in tasks where all other evaluated
algorithms fail.
Read the full abstract ▾
All authors Clarisse Wibault, Antoine Gorceix, Antonio Léon Villares, Alexey Zakharov, Evangelos Chatzaroulas, Michael Matthews, Eduardo Pignatelli, Jakob Foerster