Machine Learning
VACE: Validation-Gated Alternating Co-Evolution of Agent Models and Harnesses
Jiexing Qi, Yu He, Jun Liu et al. (+9 more)
Language model
agents can be improved by updating their model weights or refining the harness that guides task execution. These components are coupled: weight updates change how the model uses the harness, while harness updates change the trajectories used for training. We propose VACE, Validation-Gated Alternating CoEvolution, which alternates agentic reinforcement learning with trajectory-driven harness refinement. After each RL stage, VACE reuses the collected trajectories to propose a harness revision and evaluates the incumbent and candidate with the updated model held fixed. The candidate guides subsequent training only if it improves validation performance. With Qwen3.5-9B, VACE achieves 45.26% test accuracy on OfficeQA and a mean partial-credit score of 75.19% on AutomationBench, exceeding weight-only RL by 6.43 and 9.09 percentage points and ungated alternation by 4.59 and 6.95 points, respectively. Across 44 harness proposals, 17 reduce validation performance at the updated checkpoint and are rejected before subsequent RL training, highlighting the importance of validation gating.
Read the full abstract ▾
All authors Jiexing Qi, Yu He, Jun Liu, Qichen Huang, Shaohua Hu, Zhan Dang, Guohua Chen, Rui Yang, Wen Jiang, Yang Liu, Tao Lyu, Fangming Li
Computation and Language
Hongyang Li, Xiao Li, Caesar Wu et al. (+2 more)
Open-source
reasoning models provide unequal access to
reasoning capability across
languages. When a model can solve a problem but cannot deliver a complete solution in the user's
language,
language becomes an access barrier rather than merely a source of performance variation. We audit Qwen3 checkpoints on competition-mathematics tasks in eleven
languages. Across the ten non-English
languages, only 15.4-17.9% of problems receive a correct, terminating solution with visible
reasoning in the requested
language in any of 16 samples, compared with 92.9% in English. To identify the source of this disparity, we evaluate thirteen endpoints from one model family, spanning released checkpoints,
multilingual supervised fine-tuning (SFT) at two scales, controlled SFT ablations, and three reinforcement-learning (RL) reward formulations. We jointly track correctness,
language adherence, termination, and delivery efficiency. The dominant bottleneck shifts across post-training stages. Released models often reason in English.
Multilingual SFT restores target-
language reasoning, but accuracy declines across
multilingual, English-only, and single-
language SFT runs, showing that this cost is not specific to
multilingual mixing; non-English
reasoning traces additionally become prone to non-terminating loops. RL restores termination in both arms at no cost in accuracy, but only the arm whose reward includes a
language term delivers: rewarding correctness alone returns the model to English. Together, these stages establish a constructive post-training path from English-pivoted capability to
multilingual reasoning that is reliably delivered.
Read the full abstract ▾
All authors Hongyang Li, Xiao Li, Caesar Wu, Grégoire Danoy, Pascal Bouvry
Numerical Analysis
Riemannian optimization on low-rank tensor train manifolds for the Gross-Pitaevskii equation
Ivan Bioli, Carlo Marcati and Maxim Rakhuba
This work is concerned with the numerical computation of Gross-Pitaevskii ground states using the tensor train (TT) format. We regard the problem as the minimization of the Gross-Pitaevskii energy functional under a unit mass constraint, and propose a first-order Riemannian optimization scheme that preserves both the mass constraint and the low-rank representation throughout the optimization, by operating on the
manifold of unit-norm functions of fixed TT-rank. On this
manifold, we derive the energy-adaptive Riemannian gradient and show how to compute it efficiently in the TT format. Spatial discretization is performed using a spectral method with numerical integration, which allows the required computations to be carried out efficiently while preserving the low-rank format. The method is shown to substantially reduce computational time compared to full-rank computations, while maintaining accuracy for both single- and multicomponent Gross-Pitaevskii equations.
Read the full abstract ▾
Sound
Prediction-Layer Branch Calibration for Multimodal Sentiment Analysis
Yulin Sun, Kele Xu and Yong Dou
Multimodal sentiment analysis integrates textual, acoustic and visual cues, yet current
language-model-based fusion methods typically leave prediction-layer branch allocation implicit. We introduce Branch-Calibrated
Multimodal Language Fusion (BC-MLF), which explicitly models prediction-layer branch allocation through a Branch-Calibrated Task Head (BCHead), complemented by Fusion
Token Contrastive Learning (FTCL) for sentiment-aware fusion-
token regularization. FTCL organizes mean-pooled fusion-
token representations according to continuous sentiment affinity, while BCHead combines fusion, text and audiovisual predictions through a lightweight sample-adaptive constrained mixture. Without modifying the fusion backbone, BC-MLF consistently improves the reproduced DeepMLF baseline and achieves the strongest results among the compared methods on CMU-MOSEI and CH-SIMS across classification and regression metrics. The controlled ablations show that sample-adaptive prediction-layer branch allocation consistently outperforms static branch aggregation. Code is available at https://github.com/sunyulin0421/BC-MLF.
Read the full abstract ▾
Robotics
V2X-WAM: A Cooperative World Action Model for End-to-End Autonomous Driving
Junwei You, Weizhe Tang, Can Wang et al. (+6 more)
Vehicle-infrastructure cooperation can complement onboard sensing with broader and more informative observations of the traffic environment, providing valuable support for end-to-end autonomous driving. However, existing cooperative driving methods mainly exploit roadside information to enhance the representation of the current scene, while the future consequences of prospective driving actions are rarely modeled explicitly. This limits the ability of the planner to anticipate how its decisions may interact with the evolving traffic environment. To address this issue, we propose V2X-WAM, a cooperative world action model that tightly couples cooperative scene understanding, action generation, and future-world
reasoning. V2X-WAM constructs a reliability-aware spatiotemporal representation from vehicle- and infrastructure-side observations, while compressing infrastructure information into a compact quantized message for efficient communication. Based on the resulting cooperative representation, a
multimodal planner generates prospective trajectories, which explicitly condition future occupancy and dynamic-flow prediction. The predicted world consequences are then fed back to refine the planned trajectory, forming a closed interaction between action and future-world
evolution. Experiments on a large-scale real-world cooperative driving
dataset demonstrate that V2X-WAM consistently improves planning accuracy and safety over representative end-to-end cooperative driving methods, while achieving stronger future-world prediction and substantially lower communication overhead. Ablation studies further validate the effectiveness of the proposed design.
Read the full abstract ▾
All authors Junwei You, Weizhe Tang, Can Wang, Yan Zhao, Jun Hua, Haotian Shi, Wei Zhang, Lin Wang, Bin Ran
Artificial Intelligence
Breaking the Illusion of Review Reliability under Static Evaluation: SCOPE Fuzzing for LLM-based Scientific Reviewers
Zhuo Chen, Hao Zeng, Jiawei Liu et al. (+6 more)
The rapid growth of submissions and reviewing workload has accelerated the use of large
language models (LLMs) in peer review. Prior studies suggest that LLM-based reviewers can penalize content perturbations, such as overclaiming, indicating a certain degree of reliability. Yet these conclusions are largely based on a narrow set of perturbation strategies instantiated with static templates, providing limited evidence of actual reliability. In this paper, we construct a three-level evaluation framework covering perturbations to surface presentation, argumentative logic, and value judgment. Experiments on representative LLM-based reviewers reveal two limitations of static evaluation: stratified vulnerability, where perturbation effects depend on whether the paper's original review score is high or low, and perturbation undercoverage, where a single template misses vulnerabilities exposed by diverse realizations. To address these limitations, we propose SCOPE-Fuzzer, a strategy-aware fuzzer that combines feedback-driven strategy selection with adaptive mutation of paper content. By iteratively probing reviewers with dynamic perturbations, SCOPE-Fuzzer consistently uncovers vulnerabilities overlooked by static evaluation and other baselines.
Read the full abstract ▾
All authors Zhuo Chen, Hao Zeng, Jiawei Liu, Guoxiu He, Le Cai, Liu Haotan, Li Wenbo, Yong Huang, Wei Lu
Computer Vision
Why MLLMs Struggle to Count: Overcoming Individuation and Aggregation Bottlenecks with ConvStack
Liwei Che, Yihao Quan, Sen Fang et al. (+4 more)
Multimodal Large
Language Models (MLLMs) consistently struggle with fine-grained visual counting, yet the underlying causes remain poorly understood. In this work, we present a mechanistic analysis of this failure mode, identifying two critical bottlenecks inherent to the global
attention pipeline of MLLMs. First, we reveal an individuation bottleneck stemming from image patchification: because
Vision Transformers process patches independently, they struggle to group fragmented
geometric features across boundaries into distinct object representations. Second, we identify a collapse in the subsequent counting aggregation process, where representation separation rapidly diminishes as numerosity increases due to
attention compression. Identifying and formalizing these twin bottlenecks constitutes our first major contribution. To overcome them, we propose ConvStack, a lightweight architecture that operates directly in the visual
token space to explicitly aggregate and inject local spatial structures via zero-initialized residual connections. By explicitly addressing the individuation bottleneck, ConvStack provides unambiguous
geometric evidence for downstream aggregation. Remarkably, by fine-tuning exclusively on counting tasks, the model achieves substantial improvements in dense object counting and broader spatial understanding
benchmarks, without compromising on general visual capabilities.
Read the full abstract ▾
All authors Liwei Che, Yihao Quan, Sen Fang, Hongyi Wang, Ranjay Krishna, Ruixiang Tang, Vladimir Pavlovic
Multiagent Systems
LLM-Based Multi-Agent Systems over Wireless Networks: A Joint Agent–Network Design Perspective
Chao Hu, Yuan Guo, Guanlin Wu et al. (+3 more)
As large
language models (LLMs) evolve from standalone models into collaborative
agents embedded in physical systems, their
reasoning and execution are increasingly distributed across wireless edge nodes. In this setting, wireless networks are experiencing a paradigm shift from only providing data connectivity to supporting the multi-
agent reasoning workflow itself. The task performance of such network-constrained LLM-based multi-
agent systems (MASs) is jointly affected by the multi-
agent reasoning dependencies as well as the underlying network connectivity and edge resources. This coupling gives rise to various technical challenges, including the metric misalignment and message redundancy, state inconsistency and
topology mismatch, as well as resource limitation and trust discontinuity. To address these challenges, this article develops a novel joint
agent–network design perspective that coordinates decisions on both sides of the system. Specifically, we present the joint design of
agent–interaction scheduling and resource allocation, the message selection-transmission co-design, as well as the joint
agent–network
topology design and workload–resource allocation. Furthermore, we consider the network-verified provenance that is linked with
agent-side information-flow control to constrain how received information affects subsequent operations. An illustrative vehicle-to-everything (V2X) case study shows that jointly adapting
agent-side interaction decisions and network operations improves task completion under communication and edge-resource constraints, outperforming the conventional
agent-only and wireless-only separate designs.
Read the full abstract ▾
All authors Chao Hu, Yuan Guo, Guanlin Wu, Yueling Che, Han Hu, Jie Xu
Information Theory
Six Families of Binary Codes Arising from Ding's Conjectures
Xiaoqiang Wang, Shiyan Xiong, Mu yuan et al. (+3 more)
Ding \cite{Ding2016} proposed ten
conjectures on binary linear codes arising from Boolean functions. Four of them, namely
Conjectures 38–41, were subsequently proved by Göloğlu and Krasnayová \cite{GologluKrasnayova2019}. In this paper, we investigate the remaining six
conjectures, namely
Conjectures 19, 27, 30, 33, 34, and 37. For
Conjectures~19 and~27, we obtain common weight restrictions and several infinite five-weight families. For
Conjecture~30, we prove that every admissible code has three, four, or five nonzero weights, and an explicit four-weight example disproves the original “three or five weights” assertion. For
Conjecture~33, an infinite five-weight family is obtained. For
Conjecture~34, we obtain a general $(2h+1)$-weight upper bound and give an explicit six-weight counterexample, showing that the original “three or five weights” assertion is false in general, where $h$ is a positive integer. The case $h=3$ with $3\nmid m$ is also completely determined. Finally,
Conjecture~37 is completely resolved by combining the known results of Ahmadi and Shafaeiabr \cite{AhmadiShafaeiabr2023} with the treatment of the two remaining classes $(a)$ and $(b)$.
Read the full abstract ▾
All authors Xiaoqiang Wang, Shiyan Xiong, Mu yuan, Jing Qiu, Dabin Zheng, Jiawei He
Computer Vision
Task-Oriented Visual Feature Compression via Residual Vector Quantization for Device-Edge Multimodal Inference
Luning Pang, Cheng Yuan, Jiawei Shao et al. (+2 more)
Large
multimodal models (LMMs) support diverse visual understanding and
reasoning tasks but are often impractical to run entirely on resource-constrained devices. Device-edge co-inference reduces device computation, yet transmitting visual data over bandwidth-limited uplinks can introduce substantial delay. Task-oriented feature compression (TOFC) reduces the payload through feature aggregation and
entropy coding. However, continuous-feature coding remains costly, and query-agnostic aggregation may discard task-relevant local evidence. We propose query-guided task-oriented feature compression (Q-TOFC) for device-edge
multimodal inference. Q-TOFC employs residual vector quantization (RVQ) to encode each merged feature as a compact sequence of codebook indices, reducing its representation cost and allowing more features to be transmitted. It further incorporates query relevance into feature aggregation and uses a quantization error compensation adapter to mitigate the distortion introduced by discrete quantization. Experiments on seven
multimodal benchmarks show that Q-TOFC reduces the visual payload by 53.6% relative to TOFC while maintaining comparable average normalized task performance. End-to-end latency evaluations further demonstrate lower latency under bandwidth-constrained uplinks.
Read the full abstract ▾
All authors Luning Pang, Cheng Yuan, Jiawei Shao, Mingtao Huang, Yuan Shen
Computer Vision
Real2Gym: Building Gyms from Videos, Bringing Skills to Robots
Kerui Ren, Yingxiang Xu, Kaiwen Song et al. (+4 more)
Real-world
videos provide rich demonstrations of manipulation, but turning them into reusable
robot skills requires visually aligned environments, executable physical interactions, and mechanisms for learning from experience. We introduce Real2Gym, an agentic Real2Sim2Real framework that turns human and
robot demonstrations into interactive simulation gyms and brings skills acquired in simulation to physical
robots. The Real2Sim module reconstructs editable scenes, aligns objects and cameras with the input, validates demonstrated or retargeted actions through native physics execution, and generates task-conditioned variations with action-feasibility checks. Within these environments, the
agent generates executable code for manipulation stages, observes their outcomes, and distills successful attempts and failures into reusable task procedures, object-relative motions, and recovery strategies. Through a shared perception-and-control interface, these skills guide subsequent execution in simulation and on real
robots, with motions adapted to current observations and no updates to the underlying model weights. Extensive evaluations demonstrate that Real2Gym enables high-fidelity simulation environment reconstruction, outperforming GPT-6 Astra Direct Mode by 16.7% in success rate with approximately 74.9% fewer policy-execution
tokens across these environments, while exceeding it by 33.3% in physical
robot execution success rate across four tasks on a real Franka
robot.
Read the full abstract ▾
All authors Kerui Ren, Yingxiang Xu, Kaiwen Song, Linning Xu, Bo Dai, Mulin Yu, Tao Lu
Robotics
Multifunctional Locomotion Control of Multi-Jointed BURs with Swimming and gait Capabilities
Takumi Asada, Hideo Furuhashi, Kenta Tabata et al. (+2 more)
Multimodal biomimetic
underwater robots (BURs) can conduct
underwater tasks suitable for the environment. Combining the characteristics of aquatic organisms enables the swimming and leggedlocomotion required for
underwater exploration. Locomotion control mechanism relies on rule-based behavior selection and the designer's discretion. This limits the
robot's ability to acquire new behavioral capabilities to the predetermined range of behaviors. To address these challenges, we propose a mechanism and control system that enables the expression of
multimodal locomotion capabilities from the same multi-jointed structure. A mechanism equipped with four leg-fins each having four axes is used. This controller achieves nonlinear behavior based on sensor modalities, rather than relying on predefined conditional rule-based on locomotion functions. This system was validated through both physical and simulation testing based on multiple sensor data and behavioral patterns. Utilizing a potential function in
multimodal locomotion control was verified to enable transitions between two or three behaviors. Implementing the control method as a
multimodal controller is expected to enhance its application in
underwater exploration. Our project page is at https://tasada038.github.io/multi-jointed-bur/.
Read the full abstract ▾
All authors Takumi Asada, Hideo Furuhashi, Kenta Tabata, Renato Miyagusuku, Koichi Ozaki
Distributed Computing
ARGOS: Reinforcement Learning-Driven Multidimensional Elasticity for Service Orchestration in the Computing Continuum
Javier Mateos-Bravo, Sergio Laso, Juan Luis Herrera et al. (+3 more)
Data-intensive services in the Computing Continuum must balance analytics quality, resource usage, and cost across heterogeneous nodes with limited and uneven capacity. This balance becomes especially difficult when resource scaling reaches capacity limits, because changes in demand and cluster pressure must then be absorbed without violating client-defined quality ranges. Existing orchestrators mainly adapt resources, placements, or replicas, while analytics requirements such as coverage, sample, and freshness remain fixed. This article presents ARGOS, the Adaptive Reinforcement Learning-Driven Governance for Orchestrated Services, an end-to-end controller that formulates multidimensional elasticity as a per-request Markov decision process over analytics quality and cluster pressure, supported by capacity-aware admission. ARGOS is evaluated under controlled workloads and time-varying multi-tenant arrivals on a heterogeneous cluster. Across the controlled scenarios, the deep reinforcement learning policies consistently outperform the non-learning baselines and approach the independently tuned best-fixed reference. A separate live evaluation reports improvements over the static midpoint under realistic and saturated arrivals, with no recorded CPU or
memory violations but remaining coverage violations. These results support deep reinforcement learning as an adaptive mechanism for multidimensional elasticity when resource scaling alone is insufficient.
Read the full abstract ▾
All authors Javier Mateos-Bravo, Sergio Laso, Juan Luis Herrera, Ilir Murturi, Pantelis Frangoudis, Schahram Dustdar
Robotics
BCNav: Bearing-Conditioned Depth Policies for Sound Source Navigation
Yaozhong Kang, Jiang Wang, Takeshi Ashizawa et al. (+2 more)
The ability to navigate toward sound sources extends a
robot's reach beyond its visual field, enabling response to auditory events in unknown environments. To equip
robots with this capability, existing methods couple acoustic and visual information through joint audio-visual learning in acoustic simulators. However, acoustic simulation is both low-fidelity and expensive, producing a domain gap that prevents reliable real-world deployment, while the discrete action spaces inherited from grid-based simulators introduce an additional kinematic gap on physical
robots. To alleviate these issues, we propose BCNav, a decoupled framework that separates the acoustic module from the learned
navigation policy using direction-of-arrival (DOA) estimation: an
estimator provides a scalar bearing to the sound source, so the
navigation policy only processes depth images and a bearing angle, two inputs whose domain gaps are well characterized. We collect shortest-path demonstrations with calibrated bearing noise injection and train the policy via imitation learning to output continuous velocity commands directly executable on ground
robots. We demonstrate the method in simulation and on a physical
robot, navigating unknown environments without any acoustic fine-tuning, prior mapping, or real-world audio data collection. Code is available at https://github.com/york1to/bcnav.
Read the full abstract ▾
All authors Yaozhong Kang, Jiang Wang, Takeshi Ashizawa, Benjamin Yen, Kazuhiro Nakadai
Machine Learning
Identifying ODEs from Unstructured Data with Causal Representation Learning
Alessandro Trenta, Riccardo Massidda, Davide Bacciu et al. (+1 more)
We study the problem of recovering the governing ODE of a dynamical system from unstructured, high-dimensional observations such as images. Existing methods for ODE discovery typically assume direct measurements of the variables, or do not provide theoretical guarantees on the learned variables and equations. While Causal Representation Learning (CRL) methods provide guarantees on identifying variables from high-dimensional observations up to component-wise diffeomorphisms, we show that in general these variables cannot be used directly as input to equation discovery methods, which typically assume that the variables will lead to sparse equations. So we introduce SParse Equivalent Equation Discovery AutoEncoder (SPEED-AE), a framework that combines a pretrained CRL method with a component-wise autoencoder that learns transformations of variables that are amenable to sparse ODE discovery. We show that for
polynomial ODEs, this additional step allows us to restrict the identifiability of each variable from
polynomial to monomial diffeomorphisms. Experiments on Lotka-Volterra, Lorenz, and a two-pendulum system show that SPEED-AE improves on the disentanglement of the CRL methods and that it recovers ODEs that are closest to the ground truth, while achieving state-of-the-art
forecasting performance.
Read the full abstract ▾
All authors Alessandro Trenta, Riccardo Massidda, Davide Bacciu, Sara Magliacane
Computation and Language
Traverse: Learning When to Remember, Reset, and Redirect for Long-Horizon Web Search
Jingyuan Ma, Lynx Aster, He Zhang et al. (+5 more)
Long-horizon information-seeking
agents often accumulate noisy or misleading context, causing early mistakes to persist and making recovery increasingly difficult. We introduce an autonomous search harness in which the
agent manages its own search process through three states: Rubric, Answer, and Verify. The
agent first defines criteria for a valid answer, searches under these criteria, and then independently verifies the result before deciding whether to terminate or continue searching. It is further equipped with a Seal
Memory tool that enables active context management. Training this behavior with reinforcement learning, however, can induce Seal Collapse, resulting in unstable training and preventing the
agent from reliably learning when and how to use its
memory tools. We solve this with a simple strategy that trains only the final segment after context management. Our 35B model achieves 72.83 on BrowseComp, outperforming comparable open-source systems, and consistently improves over the base model across BrowseComp-ZH, xbench, DeepSearchQA, WideSearch, financial investigation, and product search. Ablations show that autonomous compression outperforms automatic compaction and validate our RL design.
Read the full abstract ▾
All authors Jingyuan Ma, Lynx Aster, He Zhang, Siyao Song, Weijie Yuan, Zhe Zhang, Kai Jia, Zhifang Sui
Computer Vision
LDM-is-AE: Latent Diffusion Model is an Auto-Encoder for End-to-End Image Generation
Zhengqiang Zhang, Lingchen Sun, Rongyuan Wu et al. (+4 more)
Latent
Diffusion Models (LDMs) typically adopt a two-stage pipeline: an auto-encoder (AE) is first pre-trained to define a latent space, then a
diffusion model is trained to perform denoising within it. Such a two-stage design introduces a representation mismatch, as the latent space is optimized for reconstruction rather than adapting the denoising dynamics. We reveal that the LDM itself is an AE, and consequently present LDM-is-AE, an end-to-end one-stage LDM training framework that eliminates the need for a separately trained tokenizer. Our key observation is that the LDM backbone actually performs a latent-to-feature-to-latent transformation at each denoising step, which can be interpreted as an internal decoding–encoding process. Leveraging this structure, we split the DiT backbone into two reciprocal components, DiT-E (i.e., DiT Encoding) and DiT-D (i.e., DiT Decoding), and impose image-space supervision on the intermediate features across all timesteps. Our model encourages the internal representation to align with the image domain throughout denoising, thereby establishing an explicit latent-to-image-to-latent path. At the zero-noise timestep, our model further performs an image-to-latent-to-image mapping, corresponding to an auto-encoding process. As a result, LDM-is-AE jointly learns latent representations and denoising dynamics in an end-to-end manner, yielding a
diffusion-native latent space tailored to the generation process. Experiments demonstrate that LDM-is-AE exhibits highly competitive generation performance, achieving an FID of 1.80 and 1.90 on 256x256 and 512x512 class-conditional image generation, respectively.
Read the full abstract ▾
All authors Zhengqiang Zhang, Lingchen Sun, Rongyuan Wu, Qiaosi Yi, Xiangtao Kong, Chaodong Xiao, Lei Zhang
Machine Learning
ZeroDiff: Zero-Shot Time Series Reconstruction via Informed-Prior Diffusion
Yingda Fan, Dan Lu and Xiaowei Jia
Time series modeling increasingly demands high-quality supervision, yet target observations remain scarce - exogenous inputs are broadly available, but target measurements are often unavailable due to cost, infrastructure, or accessibility constraints. Can models trained on observed locations reconstruct target time series where measurements have never been collected? We term this zero-shot time series reconstruction. A naive approach - directly mapping exogenous inputs to targets - can yield predictions at unobserved locations, but without target signals, such models fail to capture the intrinsic dynamics of the target variable, producing overly smooth outputs that underestimate extremes. This reveals systematic errors that call for explicit modeling and calibration. We propose ZeroDiff, which constructs an informed prior from exogenous variables alone, then learns to calibrate reconstruction errors through
diffusion - training on observed locations and generalizing to unobserved ones. Experiments across diverse real-world
datasets demonstrate significant improvements over existing approaches. Our code is available at https://github.com/YingdaFan/ZeroDiff-ICML2026.
Read the full abstract ▾
Numerical Analysis
Two-term small-time spectral expansions for controllability Gramians
Emmanuel Trélat and Enrique Zuazua
For a controllable single-input linear system in dimension $n$, the controllability Gramian
eigenvalues, ordered decreasingly, are known to have the successive small-time orders $T, T^3, \ldots, T^{2n-1}$. We refine this leading-order hierarchy by explicitly computing the next-order coefficient of every
eigenvalue and the first-order variation of its eigendirection and of the nested spectral subspaces. The resulting formulas have intrinsic expressions in the orthonormal Krylov basis:
eigenvalue corrections describe the action of the dynamics along each Krylov direction, while variations of the spectral subspaces describe coupling to the next direction. We also establish local joint real-analytic dependence on $(T, A, b)$ near $T=0$ for the
eigenvalues divided by their leading powers of $T$, consistently oriented eigenvectors, and spectral projectors. This provides convergent expansions with remainders uniform on compact families of controllable pairs. These results yield refined asymptotics for the worst-case minimum energy required to reach a unit target from the origin, the Gramian determinant and condition number, and the Ornstein–Uhlenbeck Gaussian profile in moving principal coordinates. The energy equals the reciprocal of the smallest Gramian
eigenvalue, whose eigendirection identifies the most energy-demanding targets for sufficiently small times. We further derive an exact Gramian-weighted energy identity for the forward Fokker–Planck equation and sharp anisotropic short-time smoothing asymptotics using, respectively, the Lyapunov equation and an exact
Fourier norm formula combined with the graded Gramian factorization. For symmetric dynamics, the spectral data reconstruct $A$ and determine $b$ up to its global sign.
Read the full abstract ▾
Machine Learning
UnlearningSoup: Is Repeated Tuning Necessary for Large Language Model Unlearning?
Puning Yang, Qizhou Wang, Junchi Yu et al. (+2 more)
Large
language models trained on vast corpora inherently risk memorizing harmful content that may later re-emerge in their outputs. To mitigate this issue, existing unlearning methods typically rely on training-based parameter updates, such as gradient ascent and its variants, to delete targeted content while preserving other knowledge. However, balancing the competing goals of forgetting and retention makes hyperparameter choices for these methods particularly difficult, often requiring repeated tuning to obtain a strong model that still leaves substantial room for improvement and transfers poorly across models and
datasets. To address this challenge, we investigate whether unlearning runs exhibit exploitable structure in weight space, and observe that models from different runs still lie in a shared evaluation-performance basin. This suggests that stronger models may be recovered through an unlearning-tailored soup strategy, reducing the need for repeated tuning for further improvement or new settings. Motivated by this, we propose UnlearningSoup, a unified framework that provides two strategies: EfficientSoup uses binary-search-based interpolation to quickly discover a well-performing model in the early stage, where repeated tuning would otherwise make strong model selection costly. PerformanceSoup uses reweighted souping to efficiently unlock the remaining performance potential in the later stage, where repeated tuning becomes increasingly inefficient. Extensive experiments across diverse
datasets and models show that UnlearningSoup delivers 2.4x to 3.3x efficiency gains in hyperparameter selection, while consistently improving performance across settings.
Read the full abstract ▾
All authors Puning Yang, Qizhou Wang, Junchi Yu, Bo Han, Xiuying Chen
Networking
Exponential Backoff: Meta-Stability and Implicit Admission Control
Richard Combes, Fabien Mathieu and Thomas Bonald
We analyze exponential backoff, an
algorithm used to share a single communication channel in a distributed manner between several users, in a similar way as many networking standards such as 802.11. We demonstrate that this
algorithm has a meta-stable behavior, in the sense that the system oscillates over long time-scales between meta-stable configurations where only a subset of the sources effectively access the channel. In other words, meta-stability creates a form of implicit admission control. We provide both theoretical tools and numerical experiments to understand this phenomenon further. Our experiments are fully reproducible and the code is publicly available as a Python package available at https://pypi.org/project/slotted-aloha-simulator/.
Read the full abstract ▾
Robotics
From Sky to Soil: A Morphing Aerial-Ground Robot for Seed Deployment
Namai Chandra and Lining Yao
Aerial seed broadcasting can reach remote restoration sites, but provides limited control over seed placement within the soil. This paper presents a geometry-assisted, tri-functional morphing
robot that combines aerial access, ground locomotion, and controlled-depth seed
embedding in a fly-drive-plant architecture. After landing, the platform reconfigures into a four-wheeled planting configuration: an electronically coupled dual-motor drive folds the rear arms outward to form ground wheels, while a descending front tray engages the propulsion motors with a drill gear train. Reusing the propulsion motors for drilling eliminates a dedicated drill drive. The planting sequence forms a hole, dispenses a seed, and allows the vehicle to reposition on the ground or return to its flight configuration. Ground mobility supports repeated planting without requiring a separate flight between adjacent sites. A companion controller issues reconfiguration, tray, and seed-gate commands, while a dedicated autopilot handles flight control. The morphing and planting mechanisms are validated using a hardware prototype that demonstrates ground repositioning and seed
embedding. This proof of concept establishes a hardware basis for aerial-ground seed
embedding through coordinated reconfiguration and actuator reuse.
Read the full abstract ▾
Computer Vision
Context without Commitment: Robust Dense Correspondence under Non-Rigid Deformation
Yuzhen He and Sara Homscheid
Non-rigid point-cloud registration aims to find the corresponding target point for each point on a deforming source surface. Point-level matching keeps the full target cloud available, but correspondence becomes ambiguous when different regions have similar local geometry. Regional or coarse-to-fine methods provide broader spatial context, but an incorrect regional match can exclude the correct correspondence before dense matching. We propose CoCo-Reg, which uses regional patches to enrich dense point features without allowing patch predictions to restrict the final point-level search. CoCo-Reg constructs farthest-point-sampled patches, exchanges
geometric information within and between source and target, supervises patch similarity using identity-corrected point overlap, and projects the resulting regional information back to dense point features. The final registration stage still scores the full target cloud before global point-level candidate selection. On 726 held-out ModelNet10 objects across nine deformation levels, two established learning-based baselines obtain mean correspondence errors of 0.1993 and 0.1921, whereas CoCo-Reg obtains 0.0547. Relative to its point-level baseline, this is a 72.6% reduction. CoCo-Reg achieves lower correspondence error on 92.3% of paired test objects and reduces the mean fraction of points with error above 0.1 from 47.3% to 17.3%. Chamfer distance and HD95 decrease in the same direction, and CoCo-Reg remains lower across all tested deformation levels. These results support using regional context for dense non-rigid correspondence without imposing a hard patch-level restriction on the final search. Because evaluation uses one checkpoint per method, the reported gains characterize the complete systems rather than the isolated causal contribution of an individual component. Code will be made publicly available.
Read the full abstract ▾
Robotics
Predictive Safety Curricula for Robust Legged Locomotion
Ivan Ovinnikov, Pascal Sutter, Christian Gehring et al. (+1 more)
Rare but consequential failures can persist in learned locomotion policies for legged
robots even when average task performance is high, in part because standard curricula primarily adapt task difficulty rather than the distribution of safety-critical experience. We introduce Predictive Safety Curricula (PSC), a framework for allocating locomotion training experience using learned predictions of future safety cost. PSC trains a distributional safety critic from policy rollouts and uses its predictions to prioritize both terrain contexts and previously encountered randomized events. The resulting curriculum modifies the training distribution while leaving the task reward and policy-optimization loss unchanged. We evaluate PSC in controlled rough-terrain locomotion and in production locomotion systems. PSC improves reliability relative to standard terrain progression, advantage-based replay, and learning-progress curricula, with the largest gains on difficult terrain and under degraded observations. The same allocation principle transfers to two production locomotion stacks. On ANYmal-D hardware, PSC reduces shank-collision incidence by $63\%$ relative to the learning-progress curriculum across three matched training seeds, with a reduction in every seed. On a production stair-climbing platform, PSC eliminates observed shank collisions in the evaluated hardware trials. These results show that learned predictions of future safety cost can provide an effective signal for allocating training experience toward rare failure modes and improving locomotion reliability.
Read the full abstract ▾
All authors Ivan Ovinnikov, Pascal Sutter, Christian Gehring, Jordis Herrmann
Distributed Computing
Windowed and Quantized Group-Based ADMM for Distributed Optimization in Heterogeneous Edge Networks
Gaiguo Wei, Qingying Zhang, Heqiang Wang et al. (+2 more)
Distributed optimization in edge networks is constrained by heterogeneous client computing capabilities and limited communication resources. We propose the Windowed and Quantized Group-Based Alternating Direction Method of Multipliers (WQ-GADMM) to coordinate group updates under limited activation capacity and reduce communication costs. Clients are grouped by estimated computation time. Each window activates a limited number of groups per round, and the cloud updates the global model after all groups have updated once. The method quantizes both downlink and uplink model exchanges to reduce communication costs and allows bounded model staleness and inexact proximal local updates. For smooth nonconvex objectives, we establish an average squared Karush-Kuhn-Tucker residual bound under the stated assumptions and parameter conditions. The bound consists of a term that decreases with the iteration count and a quantization-dependent error term. Experiments on MNIST and CIFAR-10 show that 12-bit communication reduces communication volume and simulated wall-clock time while maintaining test accuracy comparable to full precision. The 12-bit configuration also maintains complete group coverage and achieves shorter mean group inter-completion gaps than the evaluated baselines.
Read the full abstract ▾
All authors Gaiguo Wei, Qingying Zhang, Heqiang Wang, Yu Zhang, Xiaoxiong Zhong
Numerical Analysis
Numerical investigation of a homogenized model with effective interface conditions for Stokes flow through porous membranes
Jonas Knoch, Pietro Italiano, Markus Gahn et al. (+2 more)
We investigate numerically fluid flow through a thin porous layer, which is modelled as a lower-dimensional interface separating two bulk domains. At this interface, we impose effective transmission conditions involving homogenized coefficients rigorously derived via two-scale methods in previous works. These coefficients are determined by
cell problems encoding the microscale properties of the porous layer. In a first step, we analyze qualitatively how the effective coefficients depend on the pore geometry. Based on these computations, we then present numerical simulations of the macroscopic models incorporating the effective interface conditions. Although replacing the thin porous layer by an effective interface substantially reduces the computational cost, the resulting nonstandard transmission conditions pose additional challenges for the numerical implementation. We employ a Taylor–Hood-type mixed finite element discretization with discontinuous pressure and tangential velocity across the interface and prove stability as well as error estimates for the resulting scheme. Finally, for microscale geometries that are still computationally resolvable, we compare the results of the macroscopic models with those obtained from direct simulations of the microscopic models.
Read the full abstract ▾
All authors Jonas Knoch, Pietro Italiano, Markus Gahn, Carsten Gräser, Maria Neuss-Radu
Robotics
FACT: Fidelity-Aware Construction of Articulated Twins
Kuixiang Shao, Chuansen Nie, Yinuo Bai et al. (+2 more)
Visually plausible articulated assets may still fail during contact interactions or exhibit inaccurate motion. We present FACT (Fidelity-Aware Construction of Articulated Twins), an agentic framework that progressively constructs articulated twins to improve geometry, contact, and dynamic fidelity. The
agent drives an evidence–diagnosis–revision loop on a shared editable representation, selecting measurements and model edits using quantitative feedback, while numerical tools execute and validate the updates. It reconstructs editable articulated geometry from images through feature planning, targeted measurements, and diagnostic refinement. On this reference, it repairs collision proxies through task-aware local repartitioning before fidelity-constrained compression. Finally, it constructs response models from passive-response
videos, using simulation residuals to guide model revision and constrained physical parameter fitting. Experiments show that FACT improves
geometric reconstruction over baselines, enables more reliable interaction with simpler collision proxies, and better reproduces held-out physical responses than direct parameter inference.
Read the full abstract ▾
All authors Kuixiang Shao, Chuansen Nie, Yinuo Bai, Jiayuan Gu, Jingyi Yu
Machine Learning
Beyond Compression: Diagnosing How Post-Training Changes Mathematical Reasoning
Hongyang Li, Yiming Zhu, Xiao Li et al. (+3 more)
Post-training is central to mathematical
reasoning in modern large
language models (LLMs), but endpoint pass@1 alone underidentifies what has changed. Gains may reflect newly reachable solutions, cheaper sampling of latent solutions, surface robustness, or memorisation. We compare three post-training paths under a common diagnostic readout: our sufficiently trained off-policy
distillation trajectories, released Qwen3 off-policy-plus-on-policy
distillation endpoints, and a released DeepSeek-Math endpoint trained with Group Relative Policy Optimisation (GRPO). Our probe uses cross-surface pass@K over verbatim
prompts, paraphrases, numerical isomorphisms, and translations, plus consistency, distribution-shape, and verified supervised-fine-tuning (SFT) membership analyses. We find two regimes. On easier AMC problems, large-K ceilings are near saturation, so post-training mainly compresses sample cost. On harder AIME problems, post-training expands the large-K ceiling over the base model: sufficient off-policy
distillation already raises this ceiling, Qwen3 released endpoints raise it further, and DeepSeek-Math GRPO does not dominate sufficient off-policy
distillation at large K. English-dominant
distillation improves non-English
reasoning but preserves
language-tier gaps. A controlled-overfit audit finds limited sensitivity in current SFT-membership probes. Compression is one regime of post-training, not a universal explanation.
Read the full abstract ▾
All authors Hongyang Li, Yiming Zhu, Xiao Li, Caesar Wu, Said Mammar, Pascal Bouvry
Machine Learning
RL-PaO: Prediction as Action in Decision Making under Uncertainty
Jiahui Feng, Dafang Zhao, Zheng Chen et al. (+2 more)
Decision-making under uncertainty often relies on predicted parameters, yet accurate prediction does not necessarily lead to good operational decisions. Aligning prediction with downstream optimization requires learning from the consequences of the decisions those predictions induce. We introduce RL-PaO, a reinforcement learning framework that integrates system formulation, optimization, and decision execution into a single environment. This yields a Markov decision process in which prediction is regarded as action: it shifts the environment to produce subsequent context and reward that explicitly aligns prediction error with realized cost, and learning the optimal policy does not require differentiating through the black-box solver. We evaluate RL-PaO on day-ahead energy scheduling using real historical data. On the test year, RL-PaO achieves the lowest annual cost among the non-oracle baselines, achieving on average $10\%$ cost reduction. Moreover, RL-PaO is capable of further analyses to provide strong interpretability both from the policy
evolution perspective and the cost-accuracy trade-off.
Read the full abstract ▾
All authors Jiahui Feng, Dafang Zhao, Zheng Chen, Zhengmao Li, Lingwei Zhu
Distributed Computing
vSkipper: Translating Dynamic Layer Skipping into LLM Serving Gains
Wei Da, Yavuz Ferhatosmanoglu and Evangelia Kalyvianaki
Dynamic layer skipping reduces LLM computation by allowing each
token to execute only a subset of the model's layers. However, existing skippers rely on specialized generation loops and do not integrate with modern serving engines. As a result, fewer executed layers do not necessarily translate into lower serving latency: FlexiDepth skips 8 of Llama-3-8B's 32 layers on average, yet its standard generation loop decodes 14.6–21.0% more slowly than the base model. We present vSkipper, a virtualization layer that makes dynamic layer skippers pluggable in serving engines while preserving continuous batching, fixed-shape batches, paged KV caching, and captured decode graphs. At each routed layer, vSkipper groups
tokens by the skipper's decision and uses routed execution only when predicted to be profitable. We implement vSkipper in SGLang and evaluate the released FlexiDepth checkpoint against upstream SGLang under identical
prompts, arrivals, output lengths, and launch settings. At the knee of upstream's load curve, vSkipper reduces mean end-to-end latency by 36.8% on GSM8K and 13.6% on BBH. Under saturation, it increases request throughput by 11.3% and 7.4%. Serving adds no statistically resolved quality loss beyond the checkpoint's own. Across synthetic skip policies, two Qwen3 skippers, and three GPUs, we demonstrate reuse without workload-specific tuning. To our knowledge, vSkipper is the first system to realize serving-efficiency gains from per-
token interior layer skipping within a modern LLM serving engine. The code is open-sourced as an SGLang fork at https://github.com/AKafakA/sglang-vskipper/tree/vskipper-ref
Read the full abstract ▾