Computational Geometry
Linear-Time Transformations Between Connectivity Maintenance and Points Spreading on Linear and Cyclic Domains
Nicolás Honorato-Droguett
Given $n$ points on a line or closed cycle and a threshold $r>0$, the connectivity-maintenance problem is to move the points so that every gap between consecutive points is at most $r$, whereas the points-spreading problem requires every gap to be at least $r$. Li and Wang [CCCG 2015; CGT 2025] and Chen, Gu, Li, and Wang [SWAT 2012; DCG 2013] gave $O(n)$-time
algorithms for the cyclic versions of min-max points-spreading and min-max connectivity-maintenance, respectively. Ghadiri and Yazdanbod [CCCG 2016] gave an $O(n\log n)$-time
algorithm for the linear version of min-sum points-spreading. In this paper, we show that the two problems can be reduced in linear time to each other for both objectives and on both linear and cyclic domains. As an implication, min-sum connectivity-maintenance is solvable in $O(n\log n)$ time on both domains. Finally, we extend the reduction to points on a line with individual thresholds when their initial order is preserved.
Read the full abstract ▾
Cryptography and Security
PURA: Provably Unbiased and Robust Multi-Bit Watermarking for AI-Generated Text Attribution
Yaofei Wang, Jinyang Guo, Shuchao Du et al. (+6 more)
Fine-grained attribution of AI-generated text is becoming increasingly important for accountability and
auditing, yet existing multi-bit watermarking methods still struggle to simultaneously preserve the base generation distribution, support high-capacity payloads, and remain recoverable after editing. We present PURA, a provably unbiased and robust multi-bit watermarking method for text attribution. Instead of perturbing
token probabilities directly, PURA embeds payloads in the latent sampling space via keyed inverse transform sampling, and recovers them by treating observed
tokens as soft interval evidence and aggregating such evidence across the sequence. This design preserves the base generation distribution exactly while substantially improving recovery stability under post-editing and channel perturbations. Building on this recovery paradigm, we further develop a unified robustness analysis and show that, under bounded attack strength, the per-bit error probability decays exponentially with sequence length. Extensive experiments show that PURA substantially outperforms existing unbiased baselines in the high-payload regime. For example, when
embedding 36 bits in 200
tokens, PURA achieves a 91.7% message match rate, more than three times that of the strongest unbiased baseline, while preserving text quality and remaining statistically close to unwatermarked text, and incurring only millisecond-level verification overhead. Our code is available at
Read the full abstract ▾
All authors Yaofei Wang, Jinyang Guo, Shuchao Du, Chao Wang, Qiyi Yao, Donghui Hu, Weiming Zhang, Nenghai Yu, Kejiang Chen
Computer Vision
UR$^{2}$-MLLM: Uncertainty-aware Revisit Reasoning in Multimodal Large Language Models for Radiology Report Generation
Yucheng Chen, Yang Yu, Jiazhou Zhou et al. (+5 more)
Radiologists generate diagnostic reports through iterative and selective revisiting of suspicious regions to refine their interpretations. Recent
multimodal large
language models (MLLMs) for radiology report generation (RRG) have shifted from text-only
reasoning toward a “Thinking-with-Images” paradigm, incorporating visual evidence into the
reasoning process. However, existing methods provide static visual evidence without a dynamic revisit mechanism during
reasoning, neglecting how radiologists re-examine uncertain observations. To this end, we propose an Uncertainty-aware Revisit
Reasoning MLLM (UR$^{2}$-MLLM) framework that dynamically revisits uncertain regions during
reasoning for RRG. UR$^{2}$-MLLM is first equipped with uncertainty perception by training on an uncertainty-aware
dataset. We then construct a
multimodal reasoning trajectory
dataset together with a detect-and-copy mechanism, which guides when and where to revisit. Finally, a visual grounding reward refines this behavior through reinforcement learning, aligning the revisited regions with corresponding anatomical structures. Experiments on MIMIC-CXR and IU-Xray show that UR$^{2}$-MLLM achieves state-of-the-art performance, highlighting the value of uncertainty-aware visual revisit
reasoning for reliable and clinically aligned report generation.
Read the full abstract ▾
All authors Yucheng Chen, Yang Yu, Jiazhou Zhou, Yufei Shi, Yongying Lan, Yichi Zhang, Liyi Li, Si Yong Yeo
Game Theory
The Complexity of Minimizing Subsidies in Envy-Free House Allocation
Sijia Dai, Minming Li, Xiaowei Wu et al. (+1 more)
The house allocation problem is a classical one-sided matching problem that concerns the assignment of a set of $m$ houses to $n$
agents according to their preferences, where each
agent is assigned exactly one house. Among the various objectives studied in this setting, envy-freeness is one of the most widely adopted
fairness criteria. As envy-free house allocations do not always exist, we address this challenge by introducing subsidies and aim to compute allocations that achieve envy-freeness with minimum total subsidy. For binary instances, we show that a total subsidy of at most $(n-1)$ suffices to guarantee envy-freeness in house allocation, and this bound is tight. Building on the known NP-hardness for general utilities, we further show that computing an allocation that minimizes the total subsidy is NP-hard, even under binary utilities. However, when there are only a bounded number of types of
agents with binary utilities, the problem can be solved in
polynomial time. Finally, we present a
polynomial time
algorithm that computes the minimum subsidy required to achieve envy-freeness for two types of
agents with general utilities.
Read the full abstract ▾
All authors Sijia Dai, Minming Li, Xiaowei Wu, Yong Zhang
Computation and Language
Dual-Layer Agentic Memory with Fast Write Routing and Slow Consolidation
Wenzhi Li, Dong Nie, Rui Lan et al. (+7 more)
Large
language model (LLM)
agents operate in dynamic environments where knowledge continuously evolves. Existing
memory systems typically treat external
memory as a monotonically growing repository, inevitably leading to
retrieval degradation and increasing computational costs over time. We argue that the core challenge is not
retrieval alone, but managing the knowledge lifecycle: deciding what to externalize, update, or ultimately internalize. Inspired by Complementary Learning Systems (CLS) theory in neuroscience, we propose Dual-Layer Agentic
Memory, a framework that shifts
memory management to the write phase through cost-aware epistemic routing and periodic parametric consolidation. Incoming information is categorized as non-write, write-new, or write-update, and routed through a small-to-large model cascade that minimizes routing overhead while filtering redundant
memories. A subsequent write-back phase selectively consolidates high-value external
memories into model parameters via supervised fine-tuning. Experiments demonstrate the dual efficiency of our approach: a 1.7B/8B cascade prunes up to 68% of redundant external
memory while escalating fewer than 50% of inputs, yet retains over 98% of the downstream QA Exact Match (EM) achieved by an exhaustive retention baseline. We further show that periodic consolidation successfully internalizes external knowledge, allowing the router to adaptively suppress redundant writes as the model's epistemic boundaries evolve. Overall, our framework presents a unified paradigm for
agent memory: selective externalization followed by selective internalization. Code and
dataset will be released upon acceptance.
Read the full abstract ▾
All authors Wenzhi Li, Dong Nie, Rui Lan, Tongtong Lyu, Peiyao Wang, Lingzi Hong, Weihang Pan, Binbin Lin, Boyuan Pan, Yao Hu
Artificial Intelligence
Query-Driven Multimodal Information Extraction from Long Documents
Yikai Gao, Ding Xia and Xi Yang
In domain-specific
multimodal long documents, images and text jointly convey complex knowledge that cannot be fully captured by plain text alone. However, existing paradigms like DocVQA primarily focus on generating textual answers or localizing evidence regions, rather than outputting query-specific textual attribute values and corresponding images. To address this gap, we propose query-driven image-text joint extraction from long documents, requiring models to output query-requested textual attribute values and corresponding image bounding boxes. Based on challenges related to both user intent and document content, we designed a two-level taxonomy that operates at the query and instance levels. Further, we construct ITJoint, the first high-quality, manually annotated
benchmark for this new task, comprising 2,455 pages of domain-specific documents with numerous non-decorative images, 316 queries, and 910 answer instances. Finally, we evaluate representative standalone
Vision-
Language Models from different providers and further design Q2IT, a multi-
agent collaborative framework consisting of three progressively collaborating
agents for evidence collection, page selection, and target-image localization. Using a joint evaluation approach that assesses both text extraction and image localization, our experiments show that standalone VLMs struggle with this task, while Q2IT significantly improves performance on ITJoint, although a substantial gap remains toward perfect results.
Read the full abstract ▾
Cryptography and Security
AdaptPrint: Response-Adaptive Fingerprinting of Black-Box LLM Services
Yilin Li, Yifei Zhang and Guozhu Meng
Black-box LLM services have emerged as a practical deployment paradigm. Nevertheless, their opacity also hinders the systematic assessment of security risks and complicates copyright
auditing for model owners. Black-box LLM fingerprinting, which identifies the underlying LLM identity through query-response interactions, offers a promising way to bridge this gap. Existing approaches typically collect responses from target LLM services using a fixed set of queries and perform poorly in the presence of realistic and complex configurations (e.g., system
prompt and sampling settings). To overcome these limitations, we propose AdaptPrint, a response-adaptive fingerprinting method for revealing hidden LLM identities in black-box LLM services. AdaptPrint integrates three progressive response consistency probing strategies: Direct Probing, Continuation Probing, and Follow-up Probing. AdaptPrint determines the final LLM identity by performing similarity matching among candidate LLMs. Experimental results show that AdaptPrint significantly outperforms state-of-the-art methods among 27 candidate models, achieving Top-1, Top-3, and Top-5 accuracies of 80.6%, 90.3%, and 92.1%. AdaptPrint also demonstrates strong robustness across different defense strategies and decoding parameters.
Read the full abstract ▾
Machine Learning
Joint Causal Structure and Cluster Discovery Using Variational Inference
Avni Rajpal, Anubhav Kumar, Rishabh Karnad et al. (+2 more)
Causal discovery aims to understand the relationships between individual random variables. In many applications, such as
brain imaging and
climate modeling, it is more meaningful to consider interactions among groups of variables. Existing methods assume that knowledge of such groups or clusters is explicitly available when modeling interactions. However, in practice, these clusters as well as the causal relationships among them, are latent. In this paper, we present a novel approach based on variational inference to simultaneously infer both the latent clusters and causal structures. We learn an approximate posterior over clusters and graph-structure by considering variational distributions based on categorical and Bernoulli models respectively. We derive variational lower bounds and estimation techniques to learn variational and model parameters. The effectiveness of our proposed methods for cluster and causal discovery are demonstrated on both synthetic and real data sets.
Read the full abstract ▾
All authors Avni Rajpal, Anubhav Kumar, Rishabh Karnad, Mohammad Emtiyaz Khan, P.K. Srijith
Networking
Fleet-Scale Pod Deployment with VPC-Native Networking in Managed Kubernetes
Sri Saran Balaji Vellore Rajakumar, Jayanth Varavani, Murat Parlakisik et al. (+1 more)
Managed Kubernetes pod deployment is often presented as a choice between VPC-native and overlay networking. Public EKS documentation describes prefix-mode capacity and configuration, while prior CNI studies primarily measure steady-state throughput and latency. This paper presents an operational measurement study of fleet-scale pod deployment across three VPC-native modes on Amazon EKS: individual secondary-IP allocation, ENI preallocation, and IPv4 prefix delegation, using Cilium and Calico as overlay baselines. We compare representative mode-specific warm-pool settings, so the results describe combined operating points. At 400 nodes, the tested three-address secondary-IP warm pool required 7,403 seconds to place 80,000 pods. With the tested ENI-preallocation and prefix-delegation settings, the same workload completed in approximately 495 seconds, moving address provisioning off the pod-creation critical path. In the reported 400-node runs, prefix delegation achieved approximately 162 pods/s, within the same observed range as the fastest overlay baseline and ENI-preallocation runs, while requiring fewer successful EC2 provisioning calls per pod than individual-IP allocation by allocating addresses in /28 blocks. Because each /28 contains 16 addresses and occupies one IPv4 slot, an interface with 30 slots carries 29 prefixes, representing up to 464 potential pod addresses, compared with approximately 30 IPv4 slots per ENI for individual-IP allocation when kubelet and CNI limits are configured accordingly. Prefix delegation also has lower documented network-address usage and retains direct VPC reachability and flow-log visibility but the trade-off is that /28 allocations require contiguous subnet space.
Read the full abstract ▾
All authors Sri Saran Balaji Vellore Rajakumar, Jayanth Varavani, Murat Parlakisik, Pavani Panakanti
Software Engineering
D-Diff: An Interactive Environment for Adjusting Commit Boundaries Based on an Editable 3-way Diff
Daiki Muto, Takayoshi Ueno and Shinpei Hayashi
In version control, it is recommended that each commit include only changes related to one task. To follow this recommendation, developers may need to adjust commit boundaries, that is, to compare and modify the diffs between two consecutive commits. Existing tools either display only a single diff at a time, forcing developers to rely on their
memory when comparing diffs, or display three files simultaneously without showing the two consecutive diffs, forcing developers to infer them; both increase their cognitive load and hamper the adjustment. As a first step toward supporting this process, we propose D-Diff, an interactive diff adjustment environment for two consecutive commits that each involve the same single file. Based on a 3-way diff display, D-Diff integrates the two diffs into a single compact view, which also provides a way to modify them. A within-subjects user study with 8 participants showed that D-Diff significantly outperforms a baseline tool in terms of efficiency, reducing the median adjustment time from 530 seconds to 230 seconds (approximately 57%). D-Diff also received a higher percentage of positive responses across all usability questionnaire items, while no statistically significant difference was found in accuracy.
Read the full abstract ▾
Systems and Control · Cross-listed from math.OC
Antagonistic Control: Foundations, Scalability and Nonlinearity
Sribalaji C. Anand and André M.H. Teixeira
This paper studies the worst-case impact of constrained control inputs: an input seeks to maximize the average cost of some outputs, measured in the $L_2$ or $L_1$ norm, while remaining bounded in terms of other outputs. This problem template subsumes classical metrics such as the $H_\infty$ norm and the output-to-output gain, and arises in
adversarial control, security assessment, and robust control. For linear time-invariant systems, we provide an exact semi-definite program (SDP) when there is a single constraint, and SDPs computing upper bounds when there are multiple constraints. We derive sufficient conditions, in terms of system zeros and relative degrees, under which the worst-case cost is unbounded, together with a constructive closed-loop modification that removes the unboundedness. From a security standpoint, unbounded values reveal structural limitations in detecting certain attack inputs. For positive systems, we provide scalable formulations whose complexity grows linearly in the state dimension: a scalable SDP for quadratic costs, and an exact linear program for linear costs. The results extend to nonlinear
polynomial systems via a sum-of-squares program. We illustrate the results with numerical examples.
Read the full abstract ▾
Cryptography and Security
Lessons from the Hardware Hacking Competitions: Verification Techniques, Findings, and Insights
Sudipta Paria, Aritra Dasgupta, Raghul Saravanan et al. (+3 more)
Hardware hacking competitions have emerged as practical platforms for evaluating security weaknesses in complex System-on-Chip (SoC) designs while promoting security-aware verification and tool development. This paper presents a systematic study of SoC security verification through open-box hardware hacking competitions, focusing on practical vulnerability analysis strategies, observed findings, and lessons for security-aware verification. We present a multi-strategy vulnerability analysis methodology, combining simulation-based verification, formal verification, lint analysis, Large
Language Model (LLM)-assisted bug detection, and coverage-guided hybrid fuzzing. Representative vulnerability findings are analyzed to illustrate how different techniques expose complementary classes of security flaws, and we derive practical lessons for pre-silicon security verification. Finally, we discuss how competition
benchmarks can support the reproducible evaluation of emerging hardware security techniques and guide future security-aware EDA research.
Read the full abstract ▾
All authors Sudipta Paria, Aritra Dasgupta, Raghul Saravanan, Jayanth Thangellamudi, Sai Manoj P D, Swarup Bhunia
Machine Learning · Cross-listed from stat.ME
Efficient Regression Models for Scan Statistics
Gazi Abdur Rakib, Tristan Ashton, Ryan A. Loomis et al. (+4 more)
We introduce a new class of regression models for scan statistics on real-valued signals. These allow for improved fitting of non-stationary signals to contrast with the interval anomalies identified by the scan statistics. Our models can represent generalized likelihood ratio statistics. While these methods naively require $O(n^4)$ for a length $n$ signal, we provide algorithmic improvements which lead to linear time
algorithms (with assumptions on max interval width). Our methods, especially ones based on Nadaraya-Watson kernel regression, are demonstrated as especially effective in detecting both synthetically planted anomalies, and for identifying a real “platforming” issue in interferometric astronomy.
Read the full abstract ▾
All authors Gazi Abdur Rakib, Tristan Ashton, Ryan A. Loomis, Brian S. Mason, Eric J. Murphy, Ci Xue, Jeff M. Phillips
Systems and Control
A 2D Axisymmetric Multi-Domain DC Arc Model for Simulink Implementation
Vinod Kumar Maddineni, Rahul Reddy Devarapally, Nihar Panchal et al. (+3 more)
This study introduces a coupled multi-domain framework to simulate direct current (DC) arcs, integrating thermal, fluid dynamic, and electromagnetic phenomena. We derive a two-dimensional (2D) axisymmetric model by simplifying the magneto-hydrodynamics (MHD) equations and implement it in MATLAB/Simulink. The model captures the spatiotemporal
evolution of temperature, fluid velocity, and
magnetic fields within a wall-stabilized arc column. Simulations across a 10 A to 1500 A current range show a pronounced radial temperature gradient, with central axis peaks and rapid peripheral decay. The approach's accuracy is validated by aligning the simulated velocity and
magnetic fields with established physical paradigms. Additionally, the model accurately reproduces the inverse current-voltage characteristics observed in classical models and empirical studies, while transient analyses confirm rapid millisecond-scale voltage stabilization. This accessible computational model provides a robust foundation for investigating DC arc dynamics and facilitates system-level simulations for advanced
circuit breakers, welding technologies, and
plasma applications.
Read the full abstract ▾
All authors Vinod Kumar Maddineni, Rahul Reddy Devarapally, Nihar Panchal, NagaBabu Koganti, Addisalem Kokob W., Praveen Damacharla
Machine Learning
On the Capability Separation Between World-Model Policy Learning and Imitated World-Action Models
Yang Yu
World-action models predict a future outcome and then infer an associated action. Although this factorization can improve representation learning and data efficiency, it is unclear whether it provides stronger control capability than direct behavior cloning when both are trained from the same observational demonstrations. We compare a direct behavior-cloning policy, an imitation-trained world-action policy, and a policy optimized with an action-conditioned world model. At the controller-class level, every world-action policy can be flattened into a direct
stochastic policy with the same closed-loop trajectory distribution. At the population level, under realizability, exact optimization, common deployment information, and distribution-preserving deployment, direct behavior cloning and world-action imitation both recover the observational behavior policy. Thus, future prediction changes the learning factorization but not the unrestricted external policy class or ideal imitation target. Action-conditioned world-model learning differs by predicting outcomes under specified actions and comparing them through a control objective. We characterize the irreducible action-specific prediction error of future models that do not condition on the candidate action, identify conditions under which a world-action joint can recover an interventional forward model, and show that observational demonstrations do not identify action effects in general. Finally, we construct an environment family in which every observational learner has positive worst-case regret, whereas one informative intervention permits zero regret. The key distinction is therefore between predicting futures associated with observed behavior and predicting consequences of specified actions for policy optimization.
Read the full abstract ▾
Computation and Language · Cross-listed from eess.AS
Mitigating Speaker Leakage in Cascaded Multi-talker ASR with Diarization-based Transcript Correction
Hermann Yepdjio Nkouanga, Minwei Luo, Maggie Wigness et al. (+1 more)
While cascaded multi-talker ASR (MT-ASR) leverages state-of-the-art foundation models, its performance is often capped by speaker leakage during separation. Prior correction strategies primarily focus on lexical re-labeling for speaker attribution. We propose a complementary
pruning-based paradigm that robustly identifies and removes leakage artifacts. Our method utilizes a pre-trained speaker diarization model as a
multimodal verifier to prune transcribed segments satisfying a tripartite consensus of temporal containment, lexical cross-validation, and temporal
alignment. Results on LibriMix, LibriSpeechMix, and the AMI Meeting corpus show our
algorithm consistently reduces cpW ER across diverse overlap conditions. Specifically, on subsets with high speaker leakage, our method achieves relative cpW ER reductions of up to 29%, highlighting its effectiveness in enhancing the reliability of cascaded MT-ASR transcripts in complex acoustic environments.
Read the full abstract ▾
All authors Hermann Yepdjio Nkouanga, Minwei Luo, Maggie Wigness, Suresh Singh
Computer Vision
SAM3Dual: A 3rd Place Solution to the MOSEv2 Track, 8th LSVOS Challenge
JeongRae Kim, Chaehyun Kim and Changwon Lim
We present SAM3Dual, our third-place solution to the MOSEv2 track of the 8th Large-scale
Video Object
Segmentation (LSVOS) Challenge at ECCV 2026. SAM3Dual is a training-free inference extension of pretrained SAM 3 that explicitly separates temporal
memory into a short-term branch for recent observations and a long-term branch for interval-sampled historical representations. The two
memory responses are combined using a deterministic sequence-relative fusion schedule and conservatively modulated by the previous-frame object confidence. All pretrained SAM 3 parameters remain frozen, requiring no task-specific training, fine-tuning, test-time training, or online parameter optimization. The complete system achieved an official J&F score of 64.37 and ranked third in the MOSEv2 track. This result highlights the potential of reorganizing temporal
memory entirely at inference time to obtain competitive long-term VOS performance while preserving the pretrained model.
Read the full abstract ▾
Computation and Language
How Agents Represent Humans: Human-Directed Stereotypes in an Open Agent Social Network
Huangchen Xu, Yuan Wu and Yi Chang
LLM-based
agents are increasingly deployed in persistent social environments, where generated claims can be posted, replied to, remembered, and reused. We study human-directed stereotypes on Moltbook, an open
agent-native social platform, asking how
agents construct humans as a social category. For this human-target analysis, we introduce an annotation framework with four evaluative dimensions—morality, friendliness, competence, and autonomy—and a second-stage subtype scheme for descriptive other attributions. We find that competence dominates human-directed evaluations, while many other attributions describe humans as epistemic, cultural, or
embodied subjects. We further examine how these human representations appear in human–
agent narrative contexts and platform-level circulation. As an auxiliary comparison, we analyze
agent-internal community feedback through behavioral host affinity. Rather than reproducing the stable insider–outsider rejection often observed in human online communities, Moltbook feedback patterns are better explained by exposure, author visibility, and content selection. These findings suggest that bias in
agent societies should be studied not only as isolated model output, but also as a discourse process.
Read the full abstract ▾
Artificial Intelligence
Disagree to Explore, Agree to Commit: Routing-Guided Test-Time Scaling for Software Agents
Kang Chen, Junjie Nian, Yixin Cao et al. (+1 more)
Software-engineering
agents solve repository-level tasks through long,
stochastic tool-use trajectories, and repeated attempts often find fixes missed by one run. Test-time scaling is difficult because patches lack canonical answer forms, while sibling actions from a shared prefix are correlated. We study whether native MoE router traces can guide steering and selection without an external judge or selection-time test execution. Our analysis shows that routing provides a robust behavioral role signal;
token-granular readouts and decision-matched comparison sets turn it into effective control. We therefore introduce Risa (Routing-Informed Steering and Arbitration): within trajectories, routing encourages diverse exploration and controlled convergence during patch commitment; across separately sampled trajectories, agreement at informative patch positions selects a final candidate. We evaluate on SWE-bench Verified using open-weight sparse MoE
agents across scales and
reasoning-effort settings. Risa's routing arbitration raises the macro-average resolved rate from 44.9% under uniform sampling to 48.2% on the gpt-oss family, matching text consensus without answer-string matching, and it transfers to Qwen3.6, where it improves on uniform choice and matches text consensus on the full 500-task
benchmark.
Read the full abstract ▾
All authors Kang Chen, Junjie Nian, Yixin Cao, Yugang Jiang
Numerical Analysis
An hp-version time stepping spectral Monte Carlo method for semi-linear parabolic equations
Jiaying Feng, Zhiyuan Hui, Changtao Sheng et al. (+1 more)
In this paper, we present an $hp$-version time-stepping spectral Monte Carlo method for solving semi-linear
parabolic equations. The key innovation lies in constructing an exponentially accurate
stochastic algorithm that integrates a residual iteration scheme on Gauss-type nodes in both temporal and spatial directions with a reconstruction strategy rooted in spectral methods. To address the long-time simulations and initial singularities that are often challenging for traditional
stochastic algorithms (e.g., walk-on-spheres method), we further develop an $hp$-version time-stepping framework that employs multiple time steps and, respectively,
geometric time partitions with linearly increasing
polynomial degrees to handle these difficulties. Notably, the proposed
algorithm bypasses the need to solve linear systems required by traditional spectral methods and remarkably supports parallel computation at both temporal and spatial grid points. We rigorously establish exponential convergence rates for the multistep method within a finite number of iterations. Extensive numerical experiments are conducted to demonstrate the spectral accuracy and computational efficiency of the proposed method in long-time simulations, problems with initial singularities, and a five-dimensional problem, thereby validating the theoretical results.
Read the full abstract ▾
All authors Jiaying Feng, Zhiyuan Hui, Changtao Sheng, Chenglong Xu
Machine Learning
Unveiling the Depth-Performance Dilemma in Split-Federated Fine-tuning of LLMs
Hariharan Ramesh, Someshwaran Murugaiyan and Jyotikrishna Dass
Split
Federated Fine-tuning (SFF) is a promising paradigm for scaling Large
Language Models (LLMs) by partitioning model depth between resource-constrained clients and a centralized server. While system incentives for throughput and
privacy favor deep partitions, the impact of such configurations on model utility remains poorly understood. In this work, we identify and characterize the Depth-Performance Dilemma: the regime that maximizes system efficiency is precisely where fine-tuning quality collapses. Through a comprehensive audit across four model scales (GPT-2 to Llama-3-8B) and diverse
benchmarks, we demonstrate that deeper partitions provide monotonic gains in throughput and
privacy at the cost of catastrophic performance plateaus. We evaluate a suite of state-of-the-art
federated adapter aggregation methods including AVG, STACK, SVD, and FREEZE, revealing that while these techniques are effective in standard
Federated Learning, they fail to mitigate the artifacts unique to split architectures. Finally, we provide a mechanistic diagnosis for this failure, tracing the collapse to the near-isometric
topology of
Transformers, which allows aggregation noise to propagate without attenuation until it triggers
Attention Collapse in the server partition. Our findings challenge the prevailing assumption that partition depth is a utility-neutral tuning knob and provide a structural foundation for stable distributed LLM fine-tuning.
Read the full abstract ▾
Robotics
BehaviorWorldGen: Closing the Loop between Action Models and World Simulators via Controllable Behavior-Aware Structured World Generation
Jiaqi Wang, Zhuo Zhang, Haining Guan et al. (+15 more)
Modern driving action models are increasingly improved in a self-improvement loop, where a learned world simulator imagines future observations and the resulting data is fed back to refine the action model. However, the bottleneck of this loop lies in the simulators' inability to generate behaviorally plausible responses by surrounding
agents, making generated data both unrealistic in interaction and imbalanced in distribution. We introduce BehaviorWorldGen, a framework that closes the loop between action models and world simulators through controllable behavior-aware structured world generation. Its core component is BehaviorFlow, a meta-action-conditioned traffic-flow model that injects interpretable behavior controls and jointly generates multi-
agent rollouts. BehaviorFlow realizes the specified
agent behaviors while allowing surrounding vehicles to respond to the ego and to one another. The resulting rollouts are rendered by a world simulator into realistic multi-view observations, which are paired with corrected interaction-aware trajectories for action-model refinement. Since BehaviorWorldGen uses structured trajectories as the interface between its modules, it is compatible with diverse action models and world simulators. Experiments on world generation, scene extrapolation, and policy refinement demonstrate consistent improvements, with the largest benefits concentrated on difficult interactive scenarios.
Read the full abstract ▾
All authors Jiaqi Wang, Zhuo Zhang, Haining Guan, Tingguang Zhou, Haowen Cui, ChuanYe Wang, Zhongyang Zhu, Yulong Zheng, Xuefeng Chen, Zhen Yang, Tianchen Deng, Feiyang Tan, Hangning Zhou, Bo Dai, Lixia Shen, Xiwu Chen, Xiyang Wang, Jiajun Zhu
Sound
AudioNoisePrints: Model-free audio watermarking using spatial correlation in flow matching TTS
Timothy Tin-Long, Jian Zhu, Aidan Pine et al. (+1 more)
We present AudioNoisePrints, a training-free watermarking pipeline for flow matching and
diffusion TTS models, which requires minimal extra computation during inference and does not require retraining the TTS model or reducing the generation quality. We exploited the fact that there are strong correlations between the initial Gaussian noises and the generated outputs in
diffusion and flow matching models, such that a simple cosine correlation between the initial noise and the generated output can be used to perform watermaking. Moreover, we train a lightweight detector on top for more aggressive augmentations. Our method outperforms AudioSeal, a strong baseline for audio watermarking under strong augmentations. We experimented on F5TTS and other TTS and vocoder models, and concluded that they all exhibit similar spatial correlation properties, suggesting our watermarking scheme can be used for more flow-matching TTS models and even vocoders in the future.
Read the full abstract ▾
All authors Timothy Tin-Long, Jian Zhu, Aidan Pine, Mengzhe Geng
Databases
Residual Privacy Budgeting with Weighted Scarcity Allocation for Online Query Answering
Mina Khoshmehr and Fernando Beltran
In many practical deployments of differential
privacy, queries do not arrive all at once. We study online differentially private query answering under a finite zero-concentrated differential
privacy (zCDP) contract. In this setting, queries arrive sequentially, carry different accuracy thresholds, and may overlap with information already released. We formulate this setting as residual
privacy budgeting: for each arriving query, the mechanism first credits reusable support from previous DP outputs and then spends new budget only on the remaining support required to satisfy the current threshold. The controller separates feasible cases, where the minimal residual support is allocated exactly, from scarcity cases, where a weighted shortfall-conservation optimiser assigns limited support according to query difficulty. We define the weight using the Query Influence Factor (QIF), a diagnostic signal for query difficulty and instability rather than query importance. For scalar Gaussian exact reuse, inverse-variance fusion justifies additive support. We prove zCDP composition, residual minimality, 1-competitiveness against the offline optimum in the feasible regime, and avoidable expenditure for allocators that ignore released history. A scarcity impossibility result shows that no online allocator can guarantee a competitive ratio better than 1/n in threshold satisfaction, contextualising the QIF scarcity layer as a design choice for an inherently hard online problem.
Read the full abstract ▾
Computer Vision
VERDICT: Agreement Beats Pixel-Space Verification in Real-Document OCSR
Yani Guan, Dengpan Dong, Shuang Luo et al. (+6 more)
Optical Chemical Structure Recognition (OCSR) converts 2D molecular depictions in the published literature into SMILES, and is increasingly important for constructing large-scale chemical training
datasets. Automation at that scale requires identifying unreliable predictions in the absence of ground truth. Three families of label-free signals were compared on $263$ ACS journal depictions with verified ground truth: model confidence, re-rendering similarity, and agreement among recognizers. Pixel-space re-rendering performed little better than chance (AUROC $0.547$, $95\%$ CI $[0.465,0.629]$), and an oracle-tuned threshold on it reduced correct labels per image from $0.745$ to $0.205$. Agreement among four architecturally distinct recognizers instead reached an AUROC of $0.916$ ($[0.880,0.952]$). The two-of-four rule accepted $81.7\%$ of images at $88.8\%$ precision, the three-of-four rule $52.1\%$ at $98.5\%$. The same pattern held on CLEF-IP, UOB, and USPTO. This distinction is obscured on synthetic
benchmarks, where re-rendered predictions naturally resemble their inputs. A substance filter removed $2{,}193$ false agreements on wildcards and R-group fragments, after which the three-of-four rule rejected all $68$ generic depictions. VERDICT was then applied to PMC Open Access, producing $6{,}146$ structure labels for $4{,}833$ molecules; chemist adjudication of $400$ released labels in two independent samples yielded precisions of $0.995$ for the three-of-four tier and $0.958$ for the two-of-four tier. VERDICT therefore enables validated labels for
multimodal molecular databases linking structure images, machine-readable representations, and source-publication information. In SES AI's Molecular Universe platform, VERDICT further serves as an image-based interface for searching and retrieving molecular records.
Read the full abstract ▾
All authors Yani Guan, Dengpan Dong, Shuang Luo, Zi Wei, Joah Han, Dan Hannah, Yumin Zhang, Qichao Hu, Kang Xu
Machine Learning · Cross-listed from stat.ML
Token-Level Likelihood-Array Regression for Membership Inference and AI-Generated Text Detection
Jiajun Sun and Zhanrui Cai
Membership inference asks whether a text was used to train a
language model, whereas AI-generated text detection asks whether it was generated by a
language model rather than written by a human. Existing likelihood-based methods typically compress
token-level probabilities into a few prespecified scores, most often using only probabilities conditioned on the full preceding context. We propose likelihood-array regression (LAR), which evaluates each target
token under nested left-context windows and organizes the resulting likelihood-derived features into a structured array. After aligning arrays across texts of different lengths, LAR learns how detection information varies with context scale,
token position, and likelihood features. LAR-1 aggregates learned contributions from individual aligned
cells, while LAR-2 adds second-order features formed from pairs of evaluations of the same target
token across context lengths. For within-path quadratic model, we establish matching minimax lower and upper bounds, characterize errors from finite-dimensional approximation and random squared projections, and derive conditions under which an oracle spectral sieve attains the minimax rate. Across multiple scoring
language models, LAR substantially improves membership inference and AI-generated text detection over likelihood-based baselines. The analyses further show that shorter-context likelihoods contain information beyond conventional full-context probabilities, while second-order features provide additional gains for membership inference.
Read the full abstract ▾
Artificial Intelligence
Role-Specialized Mixture-of-Agents with Open-Weight LLMs for Clinical Prediction
Jun Hou, Yi Fang and Xuan Wang
Large
Language Models (LLMs) are increasingly applied to clinical prediction tasks such as in-hospital mortality and readmission from electronic health records (EHRs).
Privacy and compliance constraints motivate systems that can be deployed locally, which has increased interest in open-weight multi-
agent designs. However, most medical multi-
agent systems are evaluated as a single block, leaving unclear which
agent role contributes to prediction and whether
retrieval drives observed gains. We study a role-specialized Mixture-of-
Agents (MoA) that combines medical knowledge
retrieval with contrastive similar-patient
reasoning. By varying the role design while holding the
retrieval setup fixed, we localize the main effect to the final integrator. Pairing large open-weight analysts with a small open-weight integrator matches closed-model prompting on F1 for mortality prediction while flagging substantially more true high-risk patients. Mechanism analysis shows the role assignment directly yields a high-recall operating point without threshold tuning. The effect is task-dependent, with smaller gains for readmission because the available records correlate weakly with this longer-horizon outcome. These results position role design as a key factor in
privacy-constrained, training-free clinical LLM prediction.
Read the full abstract ▾
Computer Vision
When Does Visual Generation Help Visual Understanding in Unified Multimodal Models?
Yubo Zhu, Zhehan Kan, Jingyi Yang et al. (+6 more)
Unified
multimodal models (UMMs) can perform both understanding and generation, raising a central question: can visual generation improve understanding? Existing evaluations provide mixed evidence, but confound task difficulty,
reasoning paradigms, and the closed-loop interaction between generation and understanding. We introduce VGAU-Diag, a fine-grained evaluation framework for
vision generation-assisted understanding. It stratifies samples by difficulty, enables unified evaluation of multiple
reasoning paradigms, and uses Oracle-Assisted Reference Protocols. Our analysis shows that generated visual aids help on easier instances but become unreliable as
reasoning complexity increases. Oracle-assisted diagnosis further reveals that the main bottleneck often lies on the visual-understanding side rather than the visual-generation side, as current UMMs struggle to leverage even faithful visual aids. We also show that effective visual generation should target visual-understanding bottlenecks rather than add more
reasoning steps, and identify a three-stage transition from task-irrelevant noise, to misleading plausible guidance, and finally to useful assistance. These findings would be useful to guide the development of better UMMs.
Read the full abstract ▾
All authors Yubo Zhu, Zhehan Kan, Jingyi Yang, Miaolin Chen, Jinbo Xing, Kai Zhu, Zijian Wang, Sheng Zhong, Wei Tong
Cryptography and Security
MARL-Based Sequential RIS Auctions: A Physical-Layer Security Analysis
Yuanyu Zhang, Yu Zhang, Jialu He et al. (+3 more)
Reconfigurable intelligent surfaces (RISs) hold great potential to enhance coverage, spectral efficiency, and communication security by intelligently configuring their reflecting elements. When owned by a neutral RIS operator, these elements can be offered as resources for which legitimate receivers and eavesdroppers compete. This paper investigates such competition and evaluates its impact on the physical-layer security performance of legitimate receivers. To model the competition, we develop a sequential RIS auction (SRA) framework, in which a bundle of RIS elements is auctioned in each round through a first-price sealed-bid mechanism, with each bidder submitting its bid based on the achievable rate gain and remaining budget. We then formulate the sequential bidding process as a Markov game by specifying its states, actions, rewards, and state transitions. To solve the game, we propose a multi-bidder deep deterministic policy gradient (MADDPG)-based multi-bidder reinforcement learning (MARL) approach under centralized training and decentralized execution (CTDE), enabling legitimate receivers and eavesdroppers to learn bidding strategies that maximize their long-term economic surplus. Numerical results show that, under the considered eavesdropper bidding strategies, the RL-based strategy enables legitimate receivers to achieve the highest secrecy rate per unit cost, outperforming random and fixed strategies and approaching the ideal physical-layer upper bound.
Read the full abstract ▾
All authors Yuanyu Zhang, Yu Zhang, Jialu He, Zhixin Huang, Shuangrui Zhao, Yulong Shen
Artificial Intelligence
MCP-Universe RL: A Framework for Training MCP Tool-Use Agents via Reinforcement Learning
Ziyang Luo, Yan Yang, Xiangru Jian et al. (+5 more)
Reinforcement learning (RL) has become an effective way to improve the tool-use ability of large
language models (LLMs), but most existing RL frameworks stop at the policy update. For every new domain, the user is left with two hard systems problems: standing up an isolated environment for each of hundreds of concurrent trajectories and connecting it to training, and scheduling the rollout so that the GPU stays busy across long, multi-turn episodes that spend much of their time stalled on slow tool calls. We present MCP-Universe RL (MCP-U RL), an open-source framework that takes over both. It uses the Model Context Protocol (MCP) as the interface to the environment, so any tool already exposed as an MCP server plugs into training with no RL-specific integration code. It builds the two missing layers once and reuses them across domains: an environment-orchestration layer that provisions, isolates, and recycles the MCP environments over a pluggable container backend, and a rollout-orchestration layer whose staged pipeline overlaps trajectories to keep the GPU busy while episodes wait on tools. A backend-agnostic training layer then applies the update through an existing RL backend, with veRL and slime integrations. With one configuration, changing only the task specification, we train software-engineering, deep-research, and general tool-use
agents on gpt-oss-20b and improve task reward in all three.
Read the full abstract ▾
All authors Ziyang Luo, Yan Yang, Xiangru Jian, Ziji Shi, Xiaoqiang Lin, Jun Hao Liew, Silvio Savarese, Junnan Li