Computation and Language
Strong yet Controllable LLM Abstention without Learning to Abstain
Junwei Zhang, Yifei Wang, Wanli Yang et al. (+6 more)
Large
language models (LLMs) can reduce
hallucination by abstaining from questions they would likely get wrong. Abstention can be implemented outside the model (selective answering) using a confidence threshold that controls the answer rate (coverage), or learned inside it (abstention training), which can perform better but fixes coverage once trained. To compare the two, we match their coverage and decompose risk (error rate) into answering ability and distinguishing ability, i.e., how well the model answers and how accurately it abstains, respectively. The decomposition shows that abstention training performs better mainly by answering better: with strong confidence
estimators, the two have nearly equal distinguishing ability. This motivates Train for Accuracy, then Post-hoc Select (TAPS), a modular approach that keeps coverage controllable, needs only standard post-training and an off-the-shelf confidence
estimator, and benefits from advances in either component. Across three LLMs and three factual QA
benchmarks, TAPS with a probing-based confidence
estimator attains 1.5–9.4 percentage points lower mean risk than three representative abstention training methods at matched coverage. Within TAPS, accuracy training often weakens the model's own confidence signals, yet refitting the
estimator largely preserves distinguishing ability. Together, these results establish TAPS as a strong yet controllable baseline and raise the bar for future abstention training: sacrificing control over coverage should yield gains beyond what can be achieved by optimizing answering and abstaining separately.
Read the full abstract ▾
All authors Junwei Zhang, Yifei Wang, Wanli Yang, Hexiang Tan, Rui Tang, Xin Chen, Jingang Wang, Du Su, Fei Sun
Robotics
Constraint-Decomposed Impedance Control from Generative Policy for Articulated Object Manipulation
Kyungseo Park, Byeongdo Lim, Jungbin Lim et al. (+2 more)
Articulated object manipulation tasks such as opening doors, tightening bolts, and pulling drawers impose kinematic constraints on a
robot end-effector's motion that neural policies may fail to satisfy. Unlike hybrid force-position control that strictly separates force and position control into orthogonal subspaces, impedance control modulates a mechanical impedance system in the form of a virtual inertia, damping, and stiffness. However, selecting task-appropriate impedance gains remains challenging for
robot policies. In this paper, we address this problem for the case of manipulating articulated objects. Our constraint-decomposed impedance control method uses the sampling diversity of a generative policy to estimate the feasible motion directions in twist space. The key idea is to infer the feasible and constrained subspaces of twists from an ensemble of trajectories generated by the policy from the same observation. Principal component analysis (PCA) of the end-effector twists determines the feasible twist subspace, which is then used to form 6x6 impedance gain
matrices. For many common articulated objects, the feasible twist subspace is independent of the object configuration, so the subspace and resulting gains are estimated once and held fixed during manipulation, minimizing computational overhead. For less common cases where this independence does not hold, we instead estimate a configuration-independent twist subspace of higher dimension that contains the local feasible subspaces, allowing the same fixed-gain strategy to be used. Experiments across simulated and real-world single- and multi-DoF objects show that the proposed method consistently reduces constraint-violating contact loads relative to isotropic impedance control while maintaining task execution.
Read the full abstract ▾
All authors Kyungseo Park, Byeongdo Lim, Jungbin Lim, Minseok Choi, Frank C. Park
Artificial Intelligence
ReTAC: Residual Test-Time Agentic Computation for Video Spatial Reasoning
Yike Wu, Keyan Chen, Qixiong Wang et al. (+6 more)
Test-time agentic computation can improve
multimodal large
language models (MLLMs) through planning, tool use, iterative perception, and reflection, but indiscriminate computation may duplicate resolved work, introduce noisy evidence, and perturb correct predictions. We propose that such computation should be residual: allocate additional computation only to requirements unresolved by the base model and, after a localized failure, recompute only the affected states. We instantiate this formulation in ReTAC, a training-free multiagent framework for
video spatial
reasoning. ReTAC organizes supplementary computation around unresolved requirements. It plans ordered steps to acquire missing spatial evidence and combines the observations with compatible native evidence for
reasoning. In addition, it uses stage-aligned verification to localize failures to planning, perception, or
reasoning. Dependency-based recovery recomputes only affected states while reusing unaffected ones. Extensive experiments show that ReTAC achieves the highest overall score among the compared methods across three mainstream MLLMs and two widely adopted
benchmarks. Further analyses show that gains from additional computation depend not only on how much is used, but also on where it is allocated and which valid intermediate states are preserved.
Read the full abstract ▾
All authors Yike Wu, Keyan Chen, Qixiong Wang, Xi Xiang, Maofei Que, Xiaolong Jiang, Xiaoqin Zhang, Ling Shao, Shijian Lu
Robotics
Listen Before Act: Audio-Conditioned Robot Manipulation with ListenVLA
Jiaqi Fan, Rui Xu, Jiakun Xia et al. (+7 more)
Vision-
language-action (VLA) models have advanced robotic manipulation by connecting visual perception and
language instructions to action generation. However, many manipulation tasks depend on auditory cues that cannot be reliably inferred from visual observations and may need to be elicited through physical interaction. To address this gap, we present ListenVLA, an audio-conditioned VLA policy built on GR00T N1.7 for sound-guided manipulation. ListenVLA combines semantic audio
embeddings with waveform-derived acoustic cues over a causal audio history and conditions continuous action generation through audio
tokens and feature modulation. Moreover, we introduce an audio-action binding objective that uses context-matched demonstrations to link auditory differences to demonstrated action differences. The objective supervises short-horizon differences between flow-velocity predictions under different audio conditions, with shared pure-noise action inputs and fixed non-audio observations, alongside full-chunk action supervision. Furthermore, we develop LIBERO-Audio, an eight-task
benchmark covering sound-based target selection, responses to auditory events, and physical interactions that reveal acoustic evidence. Experiments on LIBERO-Audio and three real-world manipulation tasks show excellent performance. Ablation studies support the contributions of binding supervision and audio modulation, while audio swap tests show that sound influences action predictions. Our code and
videos are available at https://anonymous.4open.science/r/listenvla-F18D
Read the full abstract ▾
All authors Jiaqi Fan, Rui Xu, Jiakun Xia, Xuyu Fan, Shuailei Ma, Yuefeng Wang, Qing Wang, Jiarui Zhang, Yili Fu, Xu Li
Artificial Intelligence
Narrative Paradoxes in LLM Inference: Shaping Reasoning Trajectories toward Misalignment
Haoming Yang, Ke Ma, Yangbangyan Jiang et al. (+6 more)
Explicit
reasoning in large
language models (LLMs) is often associated with reliability, yet it can still pose risks when a model infers harmful goals under a benign context. We identify a distinct challenge: ensuring the safety of the multi-step
reasoning process itself, where locally coherent steps may still lead to globally harmful outcomes. To explore this vulnerability, we propose PDRA (Paradox-driven
Reasoning-based Attack), a novel attack method that programs a harmful
reasoning trajectory. PDRA first constructs a paradoxical narrative core by pairing a malicious actor (e.g., a terrorist) with a benign task (e.g., writing a safety bulletin). It then embeds this core into a structured analytical framework that forces the model through a mandatory three-stage analysis – surface purpose, latent intent, and operational reconstruction – guiding it to produce detailed harmful content while maintaining narrative coherence. Mechanistically, we show PDRA steers the model's internal representations away from refusal-related patterns, suppressing safety-triggering lexicons while activating planning-oriented terms. Empirically, PDRA achieves state-of-the-art attack success rates across diverse LLMs with high efficiency. \textcolor{red}{Warning: This paper may contain sensitive content.}
Read the full abstract ▾
All authors Haoming Yang, Ke Ma, Yangbangyan Jiang, Ligong Zhang, Xiaojun Jia, Longtao Huang, Yingfei Sun, Qianqian Xu, Qingming Huang
Robotics
ReAP: From Visual Failure Recognition to Action-Conditioned Risk Prediction for Runtime Failure Monitoring
Jingyuan Liu, Yunpeng Qing, Changxu Wei et al. (+1 more)
Despite rapid progress, generalist
robot policies remain brittle during execution. Reliable deployment therefore requires runtime monitoring that provides timely warnings of failure. Existing monitors commonly estimate risk from current observations or policy signals without explicitly predicting the future consequences of proposed actions. To anticipate such risks before action execution, we introduce ReAP, a future-aware runtime monitoring framework that combines visual failure recognition with action-conditioned failure prediction. Specifically, ReAP first trains a visual detector on successful and failed rollouts to learn a failure-relevant representation space and estimate current risk. An action-conditioned predictor then forecasts the resulting future representation in this space to estimate future failure risk. At runtime, the detector and predictor provide complementary signals for current-state failure recognition and future-aware risk assessment. Experiments across diverse
robot policies in simulation and the real world demonstrate that ReAP effectively balances failure-detection accuracy and warning timeliness.
Read the full abstract ▾
All authors Jingyuan Liu, Yunpeng Qing, Changxu Wei, Zhi Hou
Robotics
PACE: Temporal-Role Tactile Conditioning for Chunk-Based Insertion Policies
Zhiyuan Chen, Shixiong Xu, Shuai Chen et al. (+1 more)
Contact-rich insertion requires
robots to combine visual
alignment with tactile adjustment. However, existing visuotactile chunk-based policies collapse tactile feedback into a single generation-time fusion pathway, leaving persistent contact states, abrupt contact events, and future contact consequences implicitly modeled in a single representation. We present PACE (Phase-Aware Contact-Event Conditioning), a temporal-role tactile conditioning framework for chunk-based insertion policies. PACE organizes a shared tactile stream into three temporal roles: persistent contact states condition action-chunk generation, abrupt contact events correct the generated chunk, and contact-consequence prediction shapes action representations during training. This design preserves chunk-level action continuity while assigning distinct pathways to different temporal roles of contact feedback. Across five real-
robot insertion tasks, PACE achieves an average success rate of 90.7%, outperforming
vision-only and generation-time tactile-fusion baselines by over 44 percentage points; on the ManipulationNet
benchmark, PACE maintains 88.7% success across six geometry-clearance configurations, including an industrial-grade 0.1-mm clearance. Ablations confirm the complementary contributions of the three temporal roles, mechanistic analyses show that PACE separates normal insertion progress from contact events requiring immediate correction, and cross-signal and cross-architecture experiments indicate that this temporal-role principle extends across different physical feedback signals and policy backbones.
Videos and materials are available at https://pace-insertion.github.io/.
Read the full abstract ▾
All authors Zhiyuan Chen, Shixiong Xu, Shuai Chen, Yangwei You
Cryptography and Security
ATTCK-R: Knowledge-Aware Long-Horizon Jailbreaking for Protected LLMs
Md Mushfiqur Rahman, Tobias Rohde, Rahul Tripathi et al. (+2 more)
As large
language models are deployed behind increasingly sophisticated safety mechanisms, evaluating their robustness under realistic
adversarial conditions becomes increasingly difficult and critical. Existing jailbreaking evaluations often rely on static
prompts, limited interaction horizons, or lightly protected systems, failing to reflect protected deployment settings with
prompt-based safety protections. We show that the introduction of a
prompt-based safety protection reduces the average attack success rate of a strong multi-turn baseline from 0.896 to 0.608 (32.1% relative), demonstrating that unprotected evaluations substantially overestimate real-world vulnerabilities. We propose Adaptive Traceback and Tactic Curation for Knowledge-based Red-teaming (ATTCK-R), a planning-driven multi-turn red teaming framework designed for protected black-box systems. ATTCK-R models jailbreaking as a long-horizon trajectory optimization problem using a structured belief state, introduces optimization-based tactic curation to abstract reusable strategies from prior successes and failures, and employs informed traceback to resume attacks from strategically meaningful states rather than restarting. Evaluated across 9 widely used LLMs configured with a single
prompt-based safety protection layer, ATTCK-R consistently outperforms strong recent multi-turn jailbreaking frameworks. Ablation results confirm that long-horizon planning and tactic curation improve attack effectiveness, while adaptive traceback improves efficiency through structured recovery from conversational failure.
Read the full abstract ▾
All authors Md Mushfiqur Rahman, Tobias Rohde, Rahul Tripathi, Khalil Elkhalil, Kevin Lybarger
Social and Information Networks
HyperCD: Effective Source Detection in Hypergraphs via Causal Disentanglement
Yan Kang, Peican Zhu, Keke Tang et al. (+4 more)
Source detection in hypergraphs plays an important role in mitigating the spread of harmful rumors and ensuring social stability. Current methods frequently overlook the causal
entanglement problem in observations, thereby limiting the performance of source inference. In this work, we propose a source detection approach in Hypergraphs via Causal Disentanglement (HyperCD). To decouple causal propagation semantics from confounding noise, we propose the causal disentanglement module. A dual-branch variational encoder is utilized to disentangle snapshot
embeddings into causal and confounding latent spaces, ensuring that the inference module relies solely on causal propagation signals. To further enforce this separation, an
adversarial inference loss term is introduced to guarantee statistical independence between the decoupled branches. Moreover, to capture
diffusion dynamics in hypergraphs, an
attention-based hypergraph aggregation module is designed, which comprises the node-to-hyperedge propagation and hyperedge-to-node propagation processes. Extensive experiments on six
datasets demonstrate the superiority of our HyperCD in effectively mitigating the causal
entanglement problem.
Read the full abstract ▾
All authors Yan Kang, Peican Zhu, Keke Tang, Le Cheng, Zun Liu, Ang Dong, Yangming Guo
Computer Vision
RAR: Rate-Adaptive Rollout for Time-Consistent Interactive World Models
Yaoyang Liu, Dongyu Yan, Yihua Du et al. (+5 more)
Video-based interactive world models aim to enable real-time, user-controllable environments. Yet most existing systems are trained and rolled out at a fixed temporal rate, limiting both runtime-rate adaptation and physical-horizon coverage – a limitation we term temporal-rate coupling. Specifically, when runtime generation throughput differs from the learned temporal rate, the generated world evolves unintentionally more slowly or quickly. Moreover, using a single temporal rate imposes an inherent trade-off between temporal granularity and physical horizon: higher rates capture finer motion but cover a shorter duration under a fixed context window. To address both limitations, we introduce Rate-Adaptive Rollout (RAR), a framework that enables interactive world models to roll out at controllable temporal rates. At its core, Stride-RoPE aligns latent frames sampled at different rates on a shared physical-time coordinate, enabling consistent multi-rate rollout, unseen-rate generalization, and online rate switching without discarding history. Meanwhile, RAR uses multi-stride
video sampling to enable joint learning of fine-grained dynamics and long-horizon
evolution. Across action- and camera-controlled environments, RAR achieves temporally consistent multi-rate rollout and online rate switching while preserving high-rate generation quality and extending physical horizons under matched training budgets. RAR also transfers to pretrained world models while preserving perceptual quality.
Read the full abstract ▾
All authors Yaoyang Liu, Dongyu Yan, Yihua Du, Shengju Qian, Zeke Xie, Xin Wang, Ying-Cong Chen, Long Chen
Artificial Intelligence
FON: A Draft-Revise Dual-Organ Language Model without Sequence-Level Quadratic Attention
Helong Hu, Hongdan Pan and Shuiqing Hu
Self-
attention assigns two functions to one mechanism-where to read and how to mix-at the price of O(L^2) training, an O(T) decode state, and position signals bound to the training length. Existing
attention-free alternatives all optimize the single function of generation. We propose FON (Fusion-Organ Network), which differentiates one weight stack into a drafting organ (multi-tap learnable delays + content-adaptive recursion; causal, O(S) training, O(1) decode state, no sequence-level quadratic
attention) and a revision organ that re-processes the written prefix through multi-scale ripple kernels injected via weak gates, never generating. At 30M parameters, iso-parameter and iso-
token budget, FON reaches perplexity 131.03 on WikiText-103, beating
Transformer (197.73) and Mamba2 (152.53); the advantage persists under all depth-aligned baselines and at the 70M tier (-20.8%). At 8x length extrapolation FON degrades by ~8-9% without positional-encoding engineering, between ALiBi and RoPE. The revision channel provides a structural pathway for in-domain infilling that causal architectures lack in principle (+19% over on-demand coupling), at a perplexity cost indistinguishable within measurement precision. We report all costs and boundaries: impaired long-range
retrieval, synthetic-task disadvantages, weaker out-of-domain code infilling than a same-recipe
Transformer, and a 6.5-12.3x wall-clock gap.
Read the full abstract ▾
Computer Vision
Learning Temporal Evidence Selection for Video Anomaly Understanding
Hangil Park, Seungheon Kim and Tae-Kyun Kim
Video anomaly understanding (VAU) aims to generate
language-level interpretations of anomalous events in long
videos, requiring visual evidence that supports explanation beyond anomaly detection alone. Existing VAU pipelines commonly rely on anomaly-centric or predefined temporal sampling strategies, which efficiently emphasize suspicious moments but overlook contextual cues, object interactions, and temporally distributed evidence needed for
reasoning. Motivated by recent progress in learnable temporal selection for long-
video question answering, we revisit VAU temporal sampling as a downstream-aligned evidence selection problem. We introduce the semantic-conditioned Temporal Sampling Q-Former that learns to select temporal visual evidence through the VAU generation objective. Since generic VAU
prompts provide weak
video-specific guidance, we condition the sampler on CLIP-embedded object tags automatically extracted from the input
video. These tags are used only for temporal evidence selection and are not appended to the LLM generation
prompt, separating sampler conditioning from
prompt augmentation. The proposed sampler operates in visual-
token space, computes soft temporal selection weights from frame-level representations, and merges duplicated selections before response generation. Experiments on HIVAU-70K show that our model achieves the best performance among the compared fine-tuned models, while ablations demonstrate that learnable temporal sampling improves over fixed anomaly-focused sampling and that CLIP-embedded object-tag conditioning further improves over generic-
prompt conditioning. Our results suggest that temporal sampling in VAU should be aligned with explanation-oriented evidence utility rather than anomaly saliency alone.
Read the full abstract ▾
Artificial Intelligence
Memory-Efficient Integer-Only Training of Deep Neural Networks
Hazem Hesham Yousef Shalby, Fabrizio Pittorino and Manuel Roveri
Training neural networks in reduced precision can yield significant efficiency gains, yet all existing methods that scale to standard computer
vision benchmarks retain floating-point in at least one stage, typically the weight update or optimizer state. We present a training
algorithm that scales to modern deep neural networks, in which every tensor (weights, gradients, and activations) is stored as INT8, and every operation (forward pass, backward pass, weight update, optimizer update) is performed in integer arithmetic. Specifically, tensors are stored as signed integer mantissas paired with explicit power-of-two exponents, on which integer-only operations are performed using five custom composable building blocks (products, sums, divide-by-constant, nonlinear functions, and weight updates). This allows complete implementation of trainable integer layers, including convolutions, batch and layer normalization, softmax, and GELU. A per-group velocity mechanism with independent exponent tracking preserves gradient precision across
neurons and channels with diverse magnitudes. As every tensor fits in INT8, training
memory is roughly 4x smaller than the FP32 baseline. We validate our training
algorithm on CNNs (ResNet-18/34/50/101, MobileNetV2) and
Vision Transformers (ViT-Ti / B), demonstrating for the first time that training neural networks with INT8 tensors and integer arithmetic can achieve competitive accuracy on CIFAR-10, CIFAR-100 and ImageNet at a reduced
memory budget.
Read the full abstract ▾
Machine Learning
SLDBench v2: Active Scaling Law Discovery
Lingbei Meng, Rui Huang, Zhirong Shen et al. (+5 more)
Scaling laws predict large-model performance from small experiments, but finding a reliable law itself requires costly experiments. SLDBench tests whether
language-model
agents can discover scaling laws when all experimental results are given upfront. We introduce SLDBench v2 with two extensions. First, we expand the corpus from 8 to 127 scenarios spanning
language-model pretraining, downstream performance, and
vision-
language transfer, and select 30 challenging scenarios on which a reference power-law fit reaches $R^2 \le 0.8$; they represent at least $1.09 \times 10^{23}$ FLOPs of source-reported training compute. Second, we introduce active scaling law discovery: an
agent chooses which experiments to run, pays each one's recorded cost (e.g., training FLOPs), and revises its law as outcomes arrive, and is scored on the same hidden extrapolation targets as in the passive setting. Across 8 frontier
agent systems, median $R^2$ drops from 0.72-0.75 (passive) to 0.24-0.57 (active), and prediction error rises 1.5-4.7 times, depending on the system.
Agents favor cheap experiments, spending 8-44% of the full-pool cost on average; low-cost acquisition can yield strong predictions, but cheap purchases also leave experimental conditions needed for extrapolation unobserved. Experimental planning thus remains a key open challenge for
agents that assist model development in the prevailing AI-for-AI study.
Read the full abstract ▾
All authors Lingbei Meng, Rui Huang, Zhirong Shen, Jinxin Liu, Lingxiao Jing, Haorui Chen, Yaosong Du, Haowei Lin
Computer Vision
ClothAtlas: Joint Garment Identification and State Estimation under Self-Occlusion
Jens Lundell, Alberta Longhini, Lei Shu et al. (+6 more)
Garment manipulation requires inferring both garment identity and complete configuration from partial, self-occluded observations. This creates a chicken-and-egg problem: completing a partial observation requires a garment-specific deformation prior, while identifying the garment requires a configuration that explains the observation. We introduce ClothAtlas, which jointly identifies garments and estimates their complete configurations from partial point clouds and a library of canonical meshes. A shared graph-conditioned flow-matching model samples configurations for each candidate mesh and ranks candidates by observation fit. The model accommodates different mesh shapes, vertex counts, and connectivities, allowing new garments to be added from canonical meshes alone without retraining. On held-out DexGarmentLab garments, ClothAtlas ranks the correct mesh first among 221 candidates for all 22 test garments from a single flat observation. From a single self-occluded observation, it identifies 15/22 garments, compared with 8/22 for a deterministic graph
estimator. Persistent vertex identities support grasp-target localization on real garments, reducing mean error from 15.4 to 6.4 cm overall and from 17.0 to 8.3 cm on occluded targets compared with dense descriptor matching. Project page: https://clothatlas.github.io/
Read the full abstract ▾
All authors Jens Lundell, Alberta Longhini, Lei Shu, Hemal Arora, Evan Robert, Carlota Parés-Morlans, Marcel Büsching, Mårten Björkman, Jeannette Bohg
Computer Vision
3D Gaussian Splatting for Fine-grained Urban Scene Reconstruction via Semantic-guided Density Control and Loss Design
Yueling Liao, San Jiang, Xing Zhang et al. (+1 more)
3D Gaussian Splatting (3DGS) has been becoming a promising 3D reconstruction technique because of its advantage of high-fidelity rendering effect with high efficiency. However, it still suffers from coarse density control and semantically blind loss functions when reconstructing the fine-grained structures in complex urban scenes. Thus, this study proposes SG-GS, a semantic-guided 3DGS method for fine-grained structure reconstruction. It takes the semantic segmentation deduced by the Segment Anything Model (SAM) as a constraint and systematically integrates it into both Gaussian density control and loss function design. On the one hand, a semantic-constrained adaptive density control mechanism is constructed, which dynamically allocates gradient thresholds to different semantic regions of the scene and introduces an additional boundary-enhanced densification strategy, so that new Gaussians are directed to structurally critical regions. On the other hand, a multi-feature guided loss function is designed, and fine-grained supervision signals are gradually introduced through a phased progressive training strategy, alleviating the artifacts caused by the lack of semantic constraints in the basic loss. Experiments on three
datasets show that the proposed method consistently improves reconstruction accuracy. Compared with the original 3DGS, the PSNR increases by approximately 2.7% to 5.2%, while artifacts on linear structures and boundary blurring are effectively mitigated. The method can serve the fine-grained 3D reconstruction of urban scenes.
Read the full abstract ▾
All authors Yueling Liao, San Jiang, Xing Zhang, Zhongwen Hu
Machine Learning
Fast Quantized Johnson$-$Lindenstrauss Transforms
Angelos Korakitis and Vasileios Nakos
Quantization has emerged as one of the primary techniques for reducing the
memory footprint of large
language models during inference. A prominent example is Quantized Johnson$-$Lindenstrauss (QJL), which compresses KV caches using 1-bit asymmetric quantization while approximately preserving the inner products required by the
attention mechanism. Recent theoretical and practical work aims at replacing the dense Gaussian projections used by the aforementioned transforms with more structured
embeddings $-$ such as compositions of randomized Hadamard transforms $-$ while retaining the same
embedding dimension. In this paper, we introduce two constructions based on a composition of randomized Hadamard transforms, albeit with a Gaussian rather than Rademacher diagonal in the second stage. The first adds Gaussian dithering to smooth the quantizer, while the second inserts a random permutation to regularize the covariance structure of the resulting Gaussian measurements. We prove
embedding-dimension and running-time guarantees analogous to those of Fast JL in the non-quantized setting. The first construction achieves $m=O(\varepsilon^{-2}\log^2(d/\delta))$, while the second achieves $m=O(\varepsilon^{-2}\log(d/\delta))$ under a restriction on $m$. Both constructions can be applied in $O(d\log d)$ time.
Read the full abstract ▾
Artificial Intelligence
REVEAL: Reverse Engineering Agentic Latents from Frontier-Model Traces
Yuetian Du, Meng Wang, Yucheng Wang et al. (+9 more)
Frontier LLM
agents perform complex interactive tasks, but their inference costs motivate transferring their experience to smaller open-source
agents.
Distillation and latent communication typically require target-model training or access to source-model internals. We study how textual frontier-model trajectories can provide reusable local controls for a frozen
agent. Inspired by protocol reverse engineering, we introduce $\textit{Agentic Latents}$, comprising anchored states and dynamic profiles of feedback–response cycles elicited by replay in the target model. Our training-free framework, $\textbf{REVEAL}$ ($\textbf{REV}$erse $\textbf{E}$ngineering $\textbf{A}$gentic $\textbf{L}$atents), uses these profiles and trajectory outcomes to match and select local events from successful and failed rollouts of the same task and source model. Their response-transition contrasts yield a target-specific steering direction, applied before each response. Within-
benchmark evaluations on Terminal-Bench 2.1 and Humanity's Last Exam with tools report higher accuracy than unmodified
agents across three targets. Representation analyses and behavioral case studies support reusing frontier-model experience as local activation control.
Read the full abstract ▾
All authors Yuetian Du, Meng Wang, Yucheng Wang, Zhangzhe Chen, Wei Wang, Boyu Yang, Weiqi Zhai, Bing Zhao, Ming Kong, Zhan Zhou, Jie Liu, Qiang Zhu
Computer Vision
ProGAN-Based Synthetic Image Generation for Industrial Visual Inspection of Metallic Parts
Hossein Omid Beiki, Sepehr Asgari, Oliver Petrovic et al. (+1 more)
Automated visual inspection is essential for efficient repair and recycling of metallic components. AI-based defect detection is a strong fit. However, it is often constrained by scarce, low-diversity training data. To address this gap, we study Generative AI-driven synthetic image generation using Progressive Growing GANs (ProGAN). We implement a ProGAN pipeline with Wasserstein loss and gradient penalty and evaluate key hyperparameters on two public
datasets, MVTec AD (metal nut) and MPDD (tube). As a result, training remained stable and the model yielded realistic images. These images can augment training data, improve robustness, and reduce the cost of collecting rare defects.
Read the full abstract ▾
All authors Hossein Omid Beiki, Sepehr Asgari, Oliver Petrovic, Christian Brecher
Machine Learning · Cross-listed from stat.ML
Moment-Based Anomaly Detection in High-Dimensional Mixed Data
Davide Ferrari and Andrea Nicolodi
We propose an anomaly detection method for high-dimensional mixed-type data that maps heterogeneous features to a common scale via standardized moment-based transformations, yielding interpretable feature-level anomaly scores. These scores are combined through a dependence-aware aggregation step to construct the Mixed-data Influence (MDI) score, accounting for correlations across features and providing a unified measure of how strongly an observation departs from the reference data distribution. By incorporating higher-order moments, the MDI can detect anomalies that arise not only from location shifts, but also from changes in variability, skewness, and other distributional characteristics. The aggregation mechanism allows our method to capture both sparse anomalies concentrated in a small number of features and weak, coordinated deviations distributed across many features. We empirically investigate the distribution and thresholding behavior of the MDI score across different sample sizes, dimensions, dependence structures, and data types. Simulation experiments and two real-data applications, to multi-omics and ESG data, evaluate finite-sample calibration, detection performance, and feature-level interpretability.
Read the full abstract ▾
Networking
Island-Ready Communication Networks: Connectivity Monitoring and State Synchronization for 5G Crisis Connectivity at the Edge
Leon Janzen, Maria Koch and Matthias Hollick
Island-ready 5G deployments provide edge connectivity for isolated regions in crisis scenarios by switching to island operation when they are cut off from the global Internet. This work addresses open research challenges that hinder the practical realization of island readiness, with a focus on the required state synchronization and connectivity monitoring functionality. As a proof of concept and for performance evaluation, we integrate our solutions in open-source projects and emulate a country-wide 5G communication network with real-world cellular traffic. Our evaluation studies the switching latency of transitions to island operation, showing that most of the transition delay stems from the access network reconfiguration. We study the synchronization overhead of a highly distributed core network, finding that the synchronization traffic grows faster than linearly with the number of 5G cores, reaching peak synchronization data rates of 9.36MB/s for a deployment of 50 cores.
Read the full abstract ▾
Information Theory
Generalized Hamming Weights and Subcode Support Weight Distributions of Twisted Reed-Solomon Codes
Chao Liu, Hao Chen, Dabin Zheng et al. (+1 more)
In this paper, we determine the generalized Hamming weight hierarchies of two classes of twisted Reed–Solomon (TRS) codes. For TRS codes with one twist, we first determine all possible values of the generalized Hamming weights. We then give a complete characterization of the weight hierarchy in terms of the minimum distance of the dual code. Explicit conditions are derived for the minimum distances of the code and its dual, which make the whole weight hierarchy explicitly computable. As special cases, the obtained results recover several known characterizations for TRS codes to be MDS, NMDS codes, and $s$-gap AMDS codes. For TRS codes with two twists, we similarly determine the complete generalized Hamming weight hierarchy. In this case, the higher generalized Hamming weights are determined by the first two generalized Hamming weights of the dual code, for which explicit rank conditions are also obtained. In addition, we further determine the subcode support weight distributions of this class of single-twist TRS codes. More precisely, for each $1 \le r \le k$, we derive an explicit formula for the number of $r$-dimensional subcodes with a prescribed support size, thereby giving a complete description of the support distribution of subcodes.
Read the full abstract ▾
All authors Chao Liu, Hao Chen, Dabin Zheng, Yaqi Chen
Multiagent Systems
RelLedger: Relational Information Governance for LLM Agent Organizations
Junxi Huang, Wenshuo Chen, Nuoxuan Li et al. (+7 more)
In LLM
agent organizations, shared records determine which facts are available for later decisions. A sender consults interaction histories before deciding whether to report an event. A delivery gap across histories does not show whose relationship drove the choice or whether the resulting record matters downstream. We introduce RelLedger to audit these links. It holds the event, evidence, risk, and message fixed while crossing the sender's histories with the event target and an unrelated actor. Role reassignment and history rewrites locate the controlling cue. Message and cost interventions show when the effect appears. An attribution ledger maps each claim to its required response
cells and separates missing interventions from sampling uncertainty. Recipient replay follows delivered information into downstream decisions. Executable workflows for a simulated one-person company extend the audit to retrieved histories, tool calls, and shared ledger updates. Across GPT, Gemini, Claude, DeepSeek, and Qwen, the audit reveals model-specific control paths and smaller relation gaps under explicit loss rules. Claude sends the same factual notice 62.22 percentage points more often with conflict histories than with friendly histories. In executable workflows, strained histories reduce recipient review rates by 45.83 points relative to warm histories for Claude.
Read the full abstract ▾
All authors Junxi Huang, Wenshuo Chen, Nuoxuan Li, Kailong Wang, Meitong Shen, Yuehan Kou, Leihao Zhao, Yutao Yue, Xiucheng Wang, Nan Cheng
Artificial Intelligence
agent-risk: Risk Assessment for Changes to AI Agents
Yue Zhao, Zixiang Xu, Angela Huyue Zhang et al. (+2 more)
Before changing a deployed AI
agent (a stricter safety gate, a new model), organizations must approve the change, starting from existing logs. Extending classical risk analysis to partial records and partial changes to running systems,
agent-risk replays logged episodes, assuming nothing else changes and the
agent does not anticipate the change. The log settles episodes acting as before, a declared stop value settles stopped ones, and the rest need logged support, a saved-state rerun, or a bound. We prove what logs must record for exact replay and which saved states serve a gate family. For independent episodes, we bound the reruns a decision needs and tolerable drift from live behavior. In a bash sandbox, a 7,200-episode log alone decided neither of two retry gates, which let the
agent act after refusals the log never shows. Rerunning only each gate's blocked episodes from saved states reached both live decisions at half the
agent calls of a live test sized post hoc for them, given the log. Without per-step harm, a stopping gate's certified harm ceiling rose up to 2.5-fold, since the log hides whether harm preceded the stop. Approval thus needs evidence designed in advance: log fields, saved states, targeted reruns.
Read the full abstract ▾
All authors Yue Zhao, Zixiang Xu, Angela Huyue Zhang, S. Alex Yang, Xiyang Hu
Computer Vision
DepthTile: Depth-Guided Existing-Image Self-Tiling with Input-Side Tileability Assessment
Rui Wang, Xi Chen, Hui Dai et al. (+4 more)
Existing-image self-tiling aims to transform a given image into a seamlessly repeatable tile while preserving its original semantics and local details. This task requires weak editing to retain the source but leave periodic seams, whereas strong editing improves tileability at the large risk of content drift.Existing
diffusion-based tiling pipelines alleviate boundary discontinuities, yet lack awareness of which regions are geometrically important to preserve. In this work, we propose DepthTile, a depth-guided self-tiling method whose generation module requires no retraining of the
diffusion backbone. It converts monocular depth into a spatial preservation prior and integrates it into the timestep-dependent masking mechanism of Tiled
Diffusion. DepthTile converts depth into a spatial editability map and fuses it with the differential
diffusion mask. This strengthens source preservation in near-field regions indicated by the monocular depth prior while retaining editing flexibility near compatible background boundaries. To access when self-tiling is feasible, we introduce the Self-Tiling Ability Score (STAS), a lightweight input-side predictor based on RGB, gradient, depth, and depth-gradient cues. On a 335-image
benchmark, relative to the default Tiled
Diffusion pipeline, DepthTile reduces MAE, LPIPS, and Tiling Score by 12.8%, 26.7%, and 8.4%, respectively. STAS reliably enables difficulty-aware analysis and further supports compatibility-aware pairwise image tiling.
Read the full abstract ▾
All authors Rui Wang, Xi Chen, Hui Dai, Piaopiao Yu, Yixue Hao, Long Hu, Min Chen
Computer Vision
FlowAvatar: Real-Time Full-Body Avatars from Sparse Egocentric Inputs on Consumer XR Devices
Chenfeng Gao, Taeyoung Yeon, Sungheon Park et al. (+4 more)
Prior avatar embodiment research has studied tracking completeness (how faithfully the body, hands, and face are sensed and reproduced) and personalized appearance in isolation, each requiring specialized lab equipment. How these factors interact on consumer hardware remains an open question, because no existing system unifies body pose, face, hands, and personalized appearance on a standalone headset. We present FlowAvatar, a system enabling real-time personalized look-alike avatars with full-body pose, hand articulation, facial expression, and eye gaze tracking from sparse egocentric inputs on consumer XR devices (Meta Quest Pro). Our pose model achieves competitive spatial accuracy and state-of-the-art temporal consistency on the AMASS
benchmark against prior works with comprehensive spatial and temporal losses validated through architecture and loss ablations, while running in real-time. Two user studies (N=12, N=10) provide converging evidence that tracking completeness and personalized appearance serve complementary roles: tracking completeness broadly improves perceived motion quality, while only their combination achieves positive total embodiment.
Read the full abstract ▾
All authors Chenfeng Gao, Taeyoung Yeon, Sungheon Park, Vasco Xu, Anish Prabhu, Mar Gonzalez-Franco, Karan Ahuja
Sound
What Matters in Preference Post-Training of Continuous Autoregressive Speech Language Models
Yingda Shen, Yao Qian, Dekun Chen et al. (+6 more)
Continuous autoregressive
speech language models combine an autoregressive (AR) backbone with a
diffusion or flow-matching head. Preference post-training requires choosing how to construct pairs and which modules to update. Controlled interventions show that a perturbed acoustic patch changes subsequent backbone states, yet larger representation changes do not necessarily reduce
speech quality. We construct preference pairs through noise or inference-configuration perturbations and formulate a preference update based on reference-relative flow-matching errors. Updating the backbone with a fixed head yields stronger reward improvements than head-only adaptation in our evaluated setting. Across flow-time and module ablations, the lowest training loss does not consistently produce the best
speech. These findings support preference post-training that considers both acoustic sampling and autoregressive conditioning, with training choices validated on generated utterances.
Read the full abstract ▾
All authors Yingda Shen, Yao Qian, Dekun Chen, Yufei Liu, Junan Zhang, Chong Jing, Han Wang, Yuxiang Wang, Zhizheng Wu
Machine Learning
A Survey of Flow Matching in Reinforcement Learning
Nabuat Zaman Nahim, Fairoz Nower Khan and Peizhong Ju
Flow Matching (FM) has recently emerged as a principled and efficient generative modeling framework for reinforcement learning (RL), enabling expressive,
multimodal policy parameterizations via deterministic probability transport. Compared to
diffusion-based policies that rely on
stochastic denoising chains, FM uses sampling based on ordinary differential equations (ODEs), with learned velocity fields, which can substantially reduce inference latency and simplify the incorporation of RL objectives. As research in flow-based RL rapidly accelerates across offline continuous control, online fine-tuning, and foundation model
alignment, the literature has become highly fragmented. In this survey, we provide a comprehensive taxonomy of flow-matching approaches in reinforcement learning. We organize the literature along two axes: the target distribution being modeled (e.g., action policies, value critics, transition dynamics) and the mechanism of RL signal integration (e.g., energy-weighted regression, flow-based policy gradients, and group relative policy optimization). Furthermore, we survey emerging frontiers such as discrete and non-Euclidean action spaces, provide a systematic comparative analysis against Gaussian and
diffusion baselines, and outline critical open problems. Ultimately, this survey serves as a foundational roadmap for the next generation of generative reinforcement learning and
alignment.
Read the full abstract ▾
Computation and Language
Deep Speech Representations Track Word Timing along a Low-Dimensional Circular Trajectory
Burin Naowarat, Hao Tang and Sharon Goldwater
Speech timing is fundamental for
speech models to recognize words accurately, but how word-timing information is represented in
speech models remains underexplored. We study this question using the framework of representational geometry, which investigates whether and how concepts are represented in low-dimensional subspaces of the representation space. Prior work in cognitive neuroscience has shown that when HuBERT feature sequences are averaged across words of the same duration, they form near-circular trajectories from word onset to word offset in a two-dimensional "temporal subspace." Here we show that this circular structure of word timing is a general property of deep
speech representations, by demonstrating its emergence in nine additional
speech models covering both supervised and self-supervised. We then move beyond averaged word segments and show that in HuBERT, the circular trajectories are consistent enough at the level of individual words to perform word segmentation. Our approach, Rotary Segmentation (RoSeg), achieves 64.3% boundary F-score on Librispeech by detecting word boundaries when cycles complete in the temporal subspace. Although supervision is needed to identify a good subspace, nevertheless this result demonstrates much stronger implicit word knowledge in HuBERT than has previously been shown.
Read the full abstract ▾
Computer Vision
Graph3R: Unifying View Graph Initialization with Feed-Forward Reconstruction
Jingwen Yu, Hengli Wang, Jianhao Jiao et al. (+3 more)
Feed-forward reconstruction recovers camera poses and dense geometry from image sets whose views are assumed to come from a single scene. In practice, uncurated image collections often contain views from disjoint scenes, and reconstructing these views jointly can corrupt camera poses and geometry. Existing pipelines typically resolve view associations in separate stages before final reconstruction, leaving view graph initialization and reconstruction inference only loosely coupled. We introduce Graph3R, which unifies view graph initialization with feed-forward reconstruction through a shared
geometric representation. A lightweight verification head predicts pairwise covisibility from the multi-view features of $\pi^3$ to construct a view graph. This graph classifies the input views into scene groups. For each group, we mask the shared features before decoding camera poses and dense geometry. Reusing the multi-view features allows Graph3R to reconstruct disjoint scenes separately with a single backbone pass. Across three
benchmarks, Graph3R improves average maximum recall at 100% precision by upto 8.99 percentage points over the baseline, while improving reconstruction quality on mixed-scene inputs. Regarding the multi-session mapping task, Graph3R improves cross-session localization and achieves competitive pose estimation accuracy, while delivering a runtime speedup of approximately $35.9\times$ over the baseline when reconstructing the query group. By integrating learned view association with
geometric inference, our approach takes a step toward end-to-end reconstruction from uncurated image collections.
Read the full abstract ▾
All authors Jingwen Yu, Hengli Wang, Jianhao Jiao, Zhonghang Liu, Ping Tan, Hong Zhang