Computers and Society · Cross-listed from physics.soc-ph
Hybrid epidemic simulation framework coupling equation-based and individual-based models
Jaeyoung Kwak, Michael H. Lees, Chin Chun Ooi et al. (+1 more)
Mass gathering events (MGEs) can create intense, short-lived transmission hotspots whose effects propagate through urban populations. Quantifying these effects requires linking fine-scale interactions during gatherings with epidemic transmission across metropolitan mobility networks. We developed a hybrid multiscale simulation framework coupling empirically derived individual-based contact networks for MGEs with a
stochastic metapopulation susceptible-exposed-infected-recovered (SEIR) model. Using population and commuting data from Madrid, Spain, we investigated how event-related transmission affects epidemic dynamics and district-level invasion. Across the scenarios examined, MGEs increased epidemic peak incidence and accelerated epidemic progression, although the magnitude of these effects depended on event-specific contact patterns and background transmissibility. Dense contact patterns produced greater epidemic amplification, whereas measures reducing close-contact interactions weakened these effects. MGEs also advanced epidemic invasion across districts and increased the temporal synchronization of arrival times, producing a more synchronized metropolitan epidemic wave. The association between MGEs and earlier district-level invasion was partly explained by the size of the event-generated epidemic seed. The main patterns persisted across the epidemiological parameter ranges examined. The framework provides a multiscale approach for investigating how mass gathering management may influence metropolitan epidemic dynamics.
Read the full abstract ▾
All authors Jaeyoung Kwak, Michael H. Lees, Chin Chun Ooi, Wentong Cai
Machine Learning
Small transformers track Bayesian evidence for latent common causes via a context-invariant mechanism
Amir Mohammadpour and Michael Franke
We present an in-depth investigation of how a form of
Bayesian reasoning about common causes can emerge as a cross-contextual generalization in small, tractable
transformers. Incrementing on recent work, our set-up (i) disentangles causal mechanisms in the model from the causal structure of the true data-generating process, (ii) orients more towards natural
language prediction by considering inference of latent common causes, and (iii) considers whether and how
Bayesian evidence accumulation for latent common causes can be implemented in representations and mechanisms that allow for cross-context generalization to novel test cases.
Read the full abstract ▾
Artificial Intelligence
FONDANT: Strong and Best-Effort Planning via Antichains
Benjamin Aminof, Tuan Khai Nguyen and Sasha Rubin
A classical solution concept in fully observable nondeterministic (FOND) planning, is the strong policy (aka winning strategy in the closely related area of reactive synthesis), i.e., such a policy ensures that the goal is reached in an
adversarial environment. When strong policies are not available or there is no evidence that the environment is
adversarial, one can resort to best-effort policies, which always exist, and which follow the classic decision-theoretic principle that an
agent should not use a dominated strategy. A typical positional best-effort policy works as follows: from every state, it follows a strong policy if one exists from that state (such states are called “strong-winning”), else a weak policy if one exists from that state (“weak-winning”), and else is unconstrained (“losing”). In this work, we introduce a sound and complete planner for both best-effort planning and strong planning. The
algorithm that underpins the planner is quite simple: it represents certain sets of states, such as the winning regions, by their $\subseteq$-minimal elements. The
algorithm returns uniform policies, i.e., it returns a policy $\pi_t$ that is a strong solution starting in every strong-winning state, and it returns a policy $\pi_w$ that is a weak solution starting in every weak-winning state, and it provides a certificate for the set of losing states. We implemented the
algorithm with some simple optimizations (calling it FONDANT), and evaluated it on a
benchmark set consisting of the instances that were used in the evaluation of leading strong planners PR2 and FOND-SAT, and the best-effort planner BeSyftP. On coverage, our implementation is at least as good on all domains, and outperforms on some domains; and on wall time, it is slower on small and medium-sized instances, and outperforms on larger instances.
Read the full abstract ▾
Artificial Intelligence
PEARL: Adaptive Prefill-Decode Execution with Elasticity for Agentic Reinforcement Learning
Jiaan Zhu, Wei Gao, Youhui Bai et al. (+6 more)
Multi-turn rollout dominates the cost of agentic reinforcement learning (RL). Asynchronous execution and elastic GPU resources can accelerate this stage, but adding rollout replicas yields diminishing returns while training GPUs remain idle between updates. We observe that effective resource use also depends on the prefill–decode (PD) configuration. Both the choice between colocation and disaggregation and the optimal PD ratio vary with the workload, making resource scaling and PD configuration interdependent. Exploiting this opportunity requires selecting effective configurations and realizing their benefits within transient resource-availability windows despite reconfiguration costs. We present PEARL, an asynchronous agentic RL system that coordinates external resource elasticity, temporary reuse of idle training GPUs, and adaptive PD execution. PEARL maintains a unified GPU–worker–role state and uses runtime profiles to predict rollout batch completion time, accounting for environment-induced reductions in decode concurrency. It selects the PD mode and ratio under the current GPU budget and translates each decision into an incremental transition plan that minimizes worker and role changes. Cost-aware switching and borrowing policies suppress transitions with insufficient expected benefit while ensuring timely return of training GPUs. Our evaluation show that PEARL achieves $2.17$–$2.79\times$ the throughput of fixed-resource ROLL across different LLMs. Compared with RLBoost+, throughput improves by up to approximately 26.9% for Qwen3-8B and 36.3% for Qwen3-30B-A3B.
Read the full abstract ▾
All authors Jiaan Zhu, Wei Gao, Youhui Bai, Zewen Jin, Ju Huang, Siran Yang, Jiamang Wang, Lin Qu, Cheng Li
Cryptography and Security
Right Facts, Wrong Rule: Scope-Transfer Poisoning of Retrieval-Augmented Generation
Kaisheng Fan, Yishu Gao, Xunzhu Tang et al. (+2 more)
Retrieval-augmented generation (RAG) systems use external documents for facts and rules that guide answers. We study rule-scope misapplication, where a rule valid within its documented scope is applied to a question outside that scope. Correct facts can then support an incorrect answer even when the rule's boundary is visible. This exposes the limits of document-validity checks. We introduce LENS (Local Evidence, Non-local Steering), an LLM-
agent-based attack framework for constructing auditable instances of this failure. Its Planner identifies a grounded rule, its legitimate use, intended misuse, and document roles. Proposer and Verifier
agents construct and repair attack documents using local
retrieval and reader feedback while preserving factual support and explicit scope boundaries. Across four QA
benchmarks and three target generators, LENS achieves 47.1% target-error rate over 4,000 fixed attempts, including construction failures. A separate fixed-attempt audit yields 36.1% joint target-error rate under independent validity checks. Fact-preserving interventions link errors to rule scope. For 76.2% of target-matching default responses, the same model correctly judges the rule inapplicable when explicitly probed. With matched information and budgets, an applicability audit reduces target error from 70.8% to 16.4%, compared with 31.2% for task-
alignment checking. These findings motivate checking and enforcing rule applicability during answer generation.
Read the full abstract ▾
All authors Kaisheng Fan, Yishu Gao, Xunzhu Tang, Tegawend'e F. Bissyand'e, Weizhe Zhang
Machine Learning · Cross-listed from math.OC
Coordinated Lane-Level Variable Speed Limits and Ramp Metering for Successive Weaving Segments Considering Merging/Diverging Risks: A Hybrid Model Predictive Control and Multi-Agent Reinforcement Learning Approach
Guodong Ma, Baofeng Sun, Wenyu Yang et al. (+1 more)
Successive weaving segments (SWSs) on urban expressways are bottlenecks prone to recurrent congestion and collisions, requiring fine-grained active traffic management (ATM). Existing approaches struggle to balance the adaptive performance of data-driven optimization with the resilience and transferability of model-based control. We propose a hybrid framework to coordinate lane-level variable speed limits (VSLs) and ramp metering across SWSs. First, we reconstruct L-METANET, a lane-level macroscopic traffic flow model that captures free and forced lane changes. Second, we combine XGBoost-SHAP with a random-parameters binary logit (RPBL) model to derive analytical equations for merging and diverging collision risks and formulate system cost and reward functions. Third, we develop MPC-STMAPPO, a hierarchical controller integrating model predictive control (MPC) and multi-
agent reinforcement learning (MARL). Its upper MPC layer uses L-METANET for long-horizon rolling optimization and generates baseline commands; its lower spatiotemporal MAPPO (ST-MAPPO) layer, enhanced with Mamba
cells and graph
attention, produces residual actions for short-horizon adjustment. Real-world experiments on the 18-km Eastern Expressway in Changchun, China, show that L-METANET accurately reproduces lane-changing-induced flow redistribution and capacity drops, with state
evolution aligned with ground truth. XGBoost-SHAP-RPBL achieves AUCs above 0.80 in most tasks, outperforming conventional logit models. MPC-STMAPPO converges faster and performs better across multiple metrics than MPC- and MARL-based baselines. Under randomly fluctuating demand, it also significantly outperforms pure MARL in generalization, demonstrating strong potential for industrial deployment.
Read the full abstract ▾
All authors Guodong Ma, Baofeng Sun, Wenyu Yang, Zhihong Yao
Human-Computer Interaction
Toward a Culturally Adapted Chinese Language Agent: A Wizard-of-Oz Study of Nonverbal Behavior in Chinese-German Intercultural Interaction
Siddhant Jain, Anna Lea Reinwarth, Dimitra Tsovaltzi et al. (+2 more)
Successful intercultural communication requires more than grammatical competence. It demands sensitivity to culturally embedded social norms whose violation triggers subtle but meaningful nonverbal responses. For German learners of Mandarin Chinese, acquiring this sensitivity is critical yet poorly supported by existing
language-learning
agents. We present a Wizard-of-Oz (WoZ) study design and supporting real-time system for collecting
multimodal behavioral data from native Chinese speakers reacting to social norm violations by German learners. The system features a photorealistic MetaHuman avatar driven by Live Link face capture and MediaPipe upper-body tracking, a wizard console for real-time behavior selection, and synchronized
multimodal logging across
agent and learner streams. A layered annotation framework, based on psychological theory and covering non-observable socioemotional reactions, norm interpretation, verbal, and observable behavior thereof, and future supervision targets enables the corpus to support training of future automated cultural interpretation and behavior generation models. Four ecologically valid interaction scenarios, developed with cultural and pedagogical experts, provide the methodological and technical foundation for a culturally adapted conversational
agent for Chinese
language learning.
Read the full abstract ▾
All authors Siddhant Jain, Anna Lea Reinwarth, Dimitra Tsovaltzi, Rafael Math, Julia Renner
Artificial Intelligence
From Migration to Calibration: Aligning Agents with Deployment Requirements
Yaxiao Liu, Pengbo Liu, Yiwen Liu et al. (+3 more)
Deploying, migrating, or scaling an
agent changes its model, support system, application, and users. Calibration should align the
agent with destination requirements without requiring source imitation or universal superiority. We propose three standard families: basic capability and operational reachability; technical and professional precision, including terminology, units, reporting periods, program fidelity, and report dependencies; and user context covering roles, use cases, authorized values, permissions, and budgets. A versioned mapping distinguishes requirements to preserve, transform, add, or retire. Four instruments—
benchmarks, positive/negative examples, skills, and role-conditioned cost/utility—supply tunable reward channels within a fixed acceptance contract. Consultants annotate methods, branches, bounded loops, checkpoints, relative comparisons, and report refinement; a registry records applicable evidence. Adaptation revises configurations or trains a separate controller or skill while freezing the foundation executor. Independent hard gates govern qualification; RL and preference methods optimize candidates without supplying factual truth. Qualification and held-out improvement over the unchanged target are separate outcomes. Supplied intent labels permit a limited paired recount: small
prompt/retry gains coexist with regressions and failed gates. FinQA supplements match prior summaries and retain provenance conflicts. These observations motivate destination-specific evaluation but do not establish calibration efficacy. We specify matched-budget tests, clean reload, complete-denominator reporting, and dependency revalidation. This is a methodological proposal with partially reconciled pilot evidence; confirmatory validation and the greater-than-80% zero-shot task-context acceptance target remain outstanding.
Read the full abstract ▾
All authors Yaxiao Liu, Pengbo Liu, Yiwen Liu, Yihua Guan, Jiaxing Song, Xingjian Li
Computer Vision
Timeline-Bench: Evaluating Agents on Realistic Video-Editing Tasks, from Raw Footage to Final Cut
Gunin Gupta, Nirmit Arora and Pavan Kalyan Tankala
AI
agents increasingly carry out long-horizon professional work, but their evaluations rarely require a finished creative deliverable. To this end, we introduce Timeline-Bench, a
benchmark of 56 real
video-editing tasks, each asking an
agent to turn raw production material into a finished
video. Tasks range from selecting dialog takes and shaping interview footage into a story to cutting commercials from product shots, voiceovers and graphics. Every task provides a brief, source assets, a container and a set of tests. A task is resolved when the output passes every test. The tests check the delivery format, the content and the brief's explicit requirements, and include a quality test calibrated on 2,582 blind judgments by 43
video editors. We evaluate 16
agents that pair frontier models with coding-
agent harnesses such as Codex, Claude Code and OpenCode. The best, GPT-6 Astra in Codex with curated editorial guidance, resolves only 15 of the 56 tasks (26.8%), and the average
agent resolves 14.0%. Human editors prefer the reference edit in 83.5% of judgments. Most unresolved runs (562 of 771) fail only the quality test:
agents perceive footage through stills and transcripts and check their renders for defects, not craft. We release the tasks, verifier and per-run results at https://timelinebench.tensortest.com.
Read the full abstract ▾
Systems and Control
Strategically Robust Game-Theoretic Multi-Agent Trajectory Optimization
Victor L. Qin, Nicolas Lanzetti, Saverio Bolognani et al. (+1 more)
Aviation authorities worldwide expect Advanced Air Mobility (AAM) traffic management to be decentralized among service providers, requiring AAM flights to autonomously plan trajectories by predicting other flights' control inputs rather than relying on centralized coordination. Game-theoretic approaches that formulate multi-
agent collision avoidance as an exact dynamic potential game can efficiently find open-loop
equilibria, but they assume that
agents exactly follow their
equilibrium trajectories—an unrealistic assumption given uncertainties in actuation, perception, and computation. We propose a strategically robust formulation where each
agent protects against a fictitious adversary that, for each timestep, perturbs other
agents' control inputs within a bounded budget to minimize distance at that timestep. We show that, under reasonable assumptions on
agents' distance cost and robustness levels, the strategically robust game remains an exact dynamic potential game and admits a quasi-closed-form solution to the inner
adversarial problem for linear dynamics, which limits computational overhead. Experiments with up to eight
agents using logarithmic distance costs show that strategic robustness selects more robust trajectories in high-collision-risk configurations while leaving low-risk trajectories nearly unchanged, with only a modest increase in runtime.
Read the full abstract ▾
All authors Victor L. Qin, Nicolas Lanzetti, Saverio Bolognani, Hamsa Balakrishnan
Systems and Control
Two-Timescale Reinforcement Learning for Real-Time Optimization and Economic NMPC: Experimental Validation
Saket Adhau, Jose Matias, Sebastien Gros et al. (+1 more)
We propose a reinforcement learning (RL) framework that tunes both Real-Time Optimization (RTO) and Economic Nonlinear Model Predictive Control (ENMPC) to address plant–model mismatch in process systems. Drawing on modifier-adaptation concepts, the method parameterizes the dynamic model, stage costs, constraints, and RTO modifiers, and uses Q-learning to adjust these parameters at two timescales: a fast update for the ENMPC layer and a slow update for the RTO layer. The framework is experimentally validated on a laboratory rig emulating a three-well subsea oil-production network. Using plant measurement data, the proposed RTO-RLMPC scheme achieves 8.6% higher economic profit than nominal ENMPC, preserves input feasibility of the deployed ENMPC policy and empirically satisfies the path constraints under disturbances and measurement noise, and drives the learned model parameters toward reference plant values, independently identified from experimental data, along the directions that most affect the economic objective. This work provides one of the first experimental demonstrations of RL-tuned ENMPC integrated with RTO.
Read the full abstract ▾
All authors Saket Adhau, Jose Matias, Sebastien Gros, Sigurd Skogestad
Distributed Computing
AReaL-TIK: Stateful Agentic Optimization of Unified RL Kernels through an Optimization IR
Ran Yan, Youhe Jiang, Jiayi Nie et al. (+5 more)
Reinforcement learning (RL) post-training often uses distinct GPU kernels for rollout and policy update. In synchronous PPO and GRPO, numerical disagreement can perturb ratios between current
token probabilities and those assigned during rollout. Recomputing rollout log-probabilities with the policy-update backend avoids this discrepancy but adds a forward pass. Bitwise-consistent unified kernels permit reuse when the policy snapshot and probability processing match the objective. Their optimization must preserve agreement across distinct execution regimes. We present KernelBraid, an agentic framework starting from a hand-tuned, bitwise-consistent implementation. Its optimization intermediate representation (IR) organizes source-code search by linking implementations and modifications to numerical requirements, workload measurements, and derivation history. The
agent coordinates changes and retains verified intermediates for further exploration; promotion requires passing correctness checks and improving aggregate latency within per-workload limits. Across 12 end-to-end training configurations on H20, KernelBraid achieves 1.10x average throughput relative to AReaL with log-probability recomputation, and the mean training-reward ratio rounds to 1.00x. Isolated-layer profiling yields 1.40x average speedup in summed phase time across 15 model-GPU pairs. Operator-level evaluation covers correctness and performance for 10 operators on A100, H20, and H200, all passing the prescribed bitwise checks. Unified-
attention search achieves 2.52x speedup in summed workload latency over the starting implementation using 7M LLM
tokens; ablations assess the contributions of retained evidence and branch exploration to search efficiency and attained performance. Our code is open-sourced at https://github.com/areal-project/AReaL-TIK.
Read the full abstract ▾
All authors Ran Yan, Youhe Jiang, Jiayi Nie, Wenshuang Li, Yingqi Peng, Taiyi Wang, Tongkai Yang, Binhang Yuan
Machine Learning
CacheRepair: Learning to Repair Cross-Chunk Context in RAG for KV Cache Fusion
Genglin Wang, Wangsong Yin, Yeerzhati Abudunuer et al. (+3 more)
Multi-document
retrieval-augmented generation (RAG) requires a
language model to process multiple retrieved text chunks before answering a question. Precomputing each chunk's KV cache independently and concatenating the caches when the chunks are retrieved can accelerate this step. However, the assembled cache lacks cross-chunk
attention information, reducing answer quality. Selective recomputation methods recover the missing cross-chunk context by rerunning the target LLM on selected
tokens, incurring substantial online computation. We introduce CacheRepair, a lightweight network that learns the difference between independently computed KV caches and those produced by processing the chunks together. The network combines compressed KV features with
token embeddings and uses
attention that is bidirectional within each chunk and flows from earlier to later chunks. Each repair block receives the compressed cache features, and the predicted residual is added to every document
token's cache. Each repair network is trained for a specific frozen target LLM on a generic
retrieval corpus and reused across downstream
datasets. Our analysis shows that repair reduces KV errors both near chunk boundaries and throughout chunk interiors. Evaluation across three target LLMs and four downstream
datasets places CacheRepair on the measured answer-quality-latency Pareto frontier in eleven of twelve model-
dataset combinations. Reported time to first
token (TTFT) includes online cache transfer and repair. Across all twelve combinations, the largest repairers achieve 1.69-4.61$\times$ speedups in median TTFT over full prefill and improve mean F1 by 2.1-26.1 percentage points over direct cache reuse.
Read the full abstract ▾
All authors Genglin Wang, Wangsong Yin, Yeerzhati Abudunuer, Haoxuan Xu, Guoliang Xing, Zhenyu Yan
Machine Learning
FlexiWorld: Learning and Planning via Flexible Action Chunks Across Multiple Time Scales
Shidu Ren, Qilin Gu, Zhenghao Ni et al. (+4 more)
Latent world models predict future states for goal-directed planning using action chunks spanning multiple primitive steps. Existing methods typically use fixed-length chunks and either omit goal-conditioned action generation or limit their supervision to short goal spans. We introduce FlexiWorld, a JEPA-based world model that combines mixed-span goal supervision with variable-length action chunks to improve long-horizon control. During training, we sample varying goal spans and randomly partition the actions into variable-length chunks. We jointly train the world model with a causal action encoder that embeds variable-length chunks and an autoregressive actor that generates primitive actions sequentially. Student Forcing reduces exposure bias by training on generated action prefixes. For planning, Actor-Residual Cross-
Entropy Method (ARCEM) combines action-residual search with within-chunk autoregressive feedback and chunk-boundary latent prediction. Across four
benchmarks and goal distances, FlexiWorld with ARCEM achieves 89.29% mean success, compared with 83.98% for the strongest baseline. PushT ablations show improved direct control from mixed-span supervision, variable-length chunks, and Student Forcing. Without retraining, FlexiWorld supports different planning chunk lengths: longer chunks accelerate ARCEM by approximately $1.3\times$ on average while maintaining comparable average success.
Read the full abstract ▾
All authors Shidu Ren, Qilin Gu, Zhenghao Ni, Junhan Sun, Jiaqi Wang, Damien Scieur, Yunze Liu
Game Theory
Does AI Help or Harm? Endogenous Information Acquisition with AI Advice
Kexin Chen, Tao Lin and Jianwei Huang
AI advice increasingly enters human decision workflows before costly information acquisition and final action. This paper studies how such advice affects decision quality, measured by the gross utility of the final action. Advice shifts beliefs before the decision maker chooses an acquisition experiment with uniformly posterior-separable (UPS) costs. The analysis characterizes gross effects through the continuation gross value generated by net-optimal acquisition. Convexity of this value function is necessary and sufficient for every AI signal to weakly improve gross utility at every prior, while nonconvexity allows a binary signal to cause gross harm. The criterion gives tight state-space results. Every AI signal is weakly gross-improving in binary-state problems under arbitrary actions, payoffs, and UPS costs. A minimal three-state problem can generate gross harm through concentrated shutdown, where valuable acquisition stops. For classification problems with Shannon acquisition costs, the criterion becomes an active-set slope test. It guarantees weak gross improvement for homogeneous classification and all three-state weighted classification problems, while a minimal four-state failure arises from
attention dilution, where acquisition shifts away from high-stake distinctions. Along a Blackwell-increasing mixture path, gross utility can first decrease and then increase. With endogenous interpretation, cheaper interpretation can reduce gross utility.
Read the full abstract ▾
Data Structures and Algorithms
Low-Stretch Spanning Trees via Smoothed Analysis of Dijkstra's Algorithm
Ioannis Dorkofikis, Bernhard Haeupler, Maximilian Probst Gutenberg et al. (+3 more)
Given an undirected weighted graph $G$, a $\gamma$-approximate low-stretch spanning tree (LSST) $T \subseteq G$ is a tree that approximates the distance metric of $G$ up to a $\gamma$-factor in expectation. Currently, existing
algorithms to find a provably good LSST carefully construct an approximate shortest-path tree from an arbitrary source. The resulting
algorithms are intricate. In contrast, practitioners observed that a much simpler heuristic performs surprisingly well: choose an arbitrary root, run Dijkstra's
algorithm, and use the resulting shortest-path tree as an LSST. In this paper, we give a smoothed analysis of shortest-path tree
algorithms, such as Dijkstra's
algorithm, that explains this behavior. We show that adding a small perturbation to the weights of the input graph suffices to turn the shortest path tree rooted at an arbitrary node in the resulting graph into an $\tilde{O}(1)$-approximate LSST. We further show that the set of perturbations can be computed efficiently from few low-diameter decompositions (LDDs). Thus, our proof is also constructive in the sense of giving a novel approach to computing LSSTs.
Read the full abstract ▾
All authors Ioannis Dorkofikis, Bernhard Haeupler, Maximilian Probst Gutenberg, Antti Roeyskoe, Aurelio Sulser, Gernot Zöcklein
Computer Vision
VideoPhysEdit: Physical Counterfactual Video Editing via Rigid-Body Physical Scene Reconstruction
Conghan Yue, Yuanjie Chen, Yue Han et al. (+4 more)
Video editing has advanced substantially in recent years, with methods increasingly accounting for the visual consequences of edits, such as changes to shadows and occlusions. However, the physical consequences of edits, including changes to subsequent motion and interactions, remain less explored. We formulate this problem as physical counterfactual
video editing (PCVE), which aims to generate a counterfactual
video depicting the resulting motion and interactions given a source
video, a physical edit, and its execution frame. PCVE is challenging because it requires understanding scene physics and inferring the downstream motion and interactions induced by a physical intervention, while paired factual and counterfactual data and dedicated evaluation metrics are lacking. We introduce VideoPhysEdit, a new training-free pipeline for PCVE in rigid-body scenes. It makes physical
reasoning explicit through a novel physical scene reconstruction method that recovers a scene reproducing the observed motion and interactions under simulation, enabling the pipeline to apply physical edits as interventions and use the resulting trajectories to guide counterfactual
video generation. We further construct PCVE-RigidBench, a synthetic
benchmark with paired source and counterfactual target
videos and physical ground truth, and introduce the Physical Edit Score. VideoPhysEdit achieves substantially higher physical edit accuracy than open-source methods and commercial models while maintaining competitive visual fidelity. Its Physical Edit Score is 0.376, the only positive score among the compared methods. Qualitative comparisons on real
videos further show that VideoPhysEdit applies to real-world scenes and better depicts the downstream motion and interactions induced by the edits than the compared methods. Code: https://github.com/Hammour-steak/VideoPhysEdit
Read the full abstract ▾
All authors Conghan Yue, Yuanjie Chen, Yue Han, Ya Gao, Yunyan Xiao, WeiYao Zhang, Zhineng Chen
Machine Learning
CTP-FL: Common-Trajectory Gradient Prediction for Federated Learning
Junkang Liu
Communication-efficient
federated optimization commonly spends several gradient evaluations between server updates. Existing local-update methods use this computation to advance an independent model on each client. Under heterogeneous data, however, these models evaluate gradients at different locations, making the aggregated update difficult to interpret as a gradient of the global objective. We study an alternative use of the same computation budget: evaluate the global objective along a shared, predicted path. We propose Common-Trajectory Predictive
Federated Learning (CTP-FL). At each round, all clients construct the same sequence of query points from the current global model and the previous aggregated direction, evaluate $K$
stochastic gradients along this sequence, and upload their average. The server then performs a single global update. Thus, CTP-FL uses $K$ mini-batch gradients per client and one model-sized vector in each communication direction, matching the per-round computation and communication of full-participation FedAvg-M. Shared query points make the aggregated direction an unbiased
estimator of the average global gradient along the predicted path. The remaining discrepancy from the gradient at the current model is controlled by the path length, without assuming bounded client-gradient dissimilarity or bounded gradients. For smooth non-convex objectives, we establish an $\mathcal{O}\!\left( \sqrt{L\Delta\sigma^2/(NKR)}+L\Delta/R \right)$ average-stationarity bound under full participation. The analysis isolates a testable trade-off: extending the prediction path provides more forward-looking gradient information but increases its displacement bias.
Read the full abstract ▾
Machine Learning
Explaining Hyperbolic Neural Networks via Geometry-Aware Relevance Propagation
Ping Xiong, Shanglin Li, Yi Ding et al. (+2 more)
Hyperbolic neural networks introduce
geometric operations that require explicit treatment in relevance propagation. Equivalent
geometric realizations can produce different feature attributions, even when local relevance is conserved. We study this problem through
Geometric Representation Invariance (GRI), a specialization of Implementation Invariance, and zero-
curvature consistency, which requires identity relevance propagation when a
geometric module approaches the identity. We propose LRP-radial-all for origin-centered radial modules, treating
geometric scaling as modulation and assigning relevance entirely to the signal branch. The rule conserves relevance, is invariant to equivalent radial factorizations, and satisfies zero-
curvature consistency, yielding GRI for a specified Poincaré-Lorentz logarithmic-map construction. In contrast, a conservative LRP-half baseline can violate both consistency criteria. Experiments on
hyperbolic MNIST, sEEG, and CIFAR-10 classifiers assess attribution fidelity, qualitative explanations, and runtime. LRP-radial-all achieves competitive attribution fidelity across
datasets with runtime comparable to Gradient$\times$Input and substantially lower than Integrated Gradients. These findings motivate geometry-aware propagation rules that distinguish relevance conservation from consistency across equivalent computations.
Read the full abstract ▾
All authors Ping Xiong, Shanglin Li, Yi Ding, Thomas Schnake, Shinichi Nakajima
Machine Learning
A Multimodal Autonomic Sensing Framework for Objective Assessment of Patient Responses to Dental Pulp Stimulation
Youngsun Kong, Yubin Choi, Dongjin Song et al. (+3 more)
Patient responses to dental pulp testing, ranging from no sensation to intense pain, provide important information for assessing pulp status in endodontic diagnosis. However, pain is a subjective sensory and emotional experience that varies considerably across individuals and can be difficult to communicate. We investigated whether complementary autonomic signals could support objective assessment of responses during dental examination. Forty-nine patients underwent cold pulp testing, yielding no-response, mild-response, and intense-response conditions. The framework integrated ECG-derived skin nerve activity (SKNA) and R-R intervals (RRI), together with electrodermal activity (EDA), using temporal convolutional network encoders with
attention-based mid-level fusion. Individual baseline signals and subject-level
covariates, including anxiety scores and biological sex, were also incorporated. The framework achieved 80.2% balanced accuracy, 75.2% sensitivity, and 85.2% specificity for binary classification of no response versus mild or intense response. For three-class classification, it achieved 60.0% balanced accuracy and a 58.8% macro-averaged F1 score. Ablation and
attention-weight analyses indicated that EDA contributed most strongly to model performance, followed by RRI, while SKNA improved balanced accuracy by approximately five percentage points. Age was significantly associated with model performance. These findings support the feasibility of
multimodal autonomic sensing for objective, non-invasive assessment of responses to dental pulp stimulation.
Read the full abstract ▾
All authors Youngsun Kong, Yubin Choi, Dongjin Song, Dong-Guk Shin, I-Ping Chen, Ki Chon
Computer Vision
Style-Driven Data Synthesis and Degradation-Aware Enhancement for Ultrasound Image Restoration
Yu-Kai Wang, Chun-Xin Tan, Manh-Hung Nguyen et al. (+1 more)
Low-cost handheld ultrasound devices can be widely deployed compared to professional hospital ultrasound machines. However, their images suffer from compound degradation that can mislead clinical judgment. Motivated by this observation, mapping handheld low-quality (LQ) to hospital high-quality (HQ) images has been considered a valuable research question. Conventionally, the mapping requires pixel-aligned LQ-HQ pairs. This requirement is unsatisfactory in practical scenarios because real scans at different times are never pixel-aligned. This paper addresses the challenge with a two-stage framework. The first stage generates pixel-aligned LQ-HQ
datasets, and the second stage trains an enhancement model that improves LQ images. The first stage trains a cycle-consistent style-transfer model on unaligned real LQ-HQ pairs to learn a HQ-to-LQ model. Then, the model transforms real HQ images into pixel-aligned LQ images. Based on the
dataset generated by the first stage, the second stage uses the Dual Degradation-Guided (DDG) Low-Rank Adaptation (LoRA) method to fine-tune an LQ-to-HQ model based on aligned pairs. In this stage, the model is based on the well known PiSA-SR framework but inserts a degradation-conditioned correction
matrix. Experimental results on the USenhance2023
dataset show that the FID metric is improved by 16.7% over the strongest baseline while other metrics indicate that our enhanced outputs are well aligned with the real HQ distribution. The source code of our method is available at https://github.com/Jason0411202/DDG_LoRA.
Read the full abstract ▾
All authors Yu-Kai Wang, Chun-Xin Tan, Manh-Hung Nguyen, Ching-Chun Huang
Sound
RemixIT-TSE: Progressive Synthetic-to-Real Adaptation for Target Speech Extraction via Target-Aware Supervision and Remixing
Yu Wang, Haixin Guan, Shuang Wei et al. (+1 more)
Target
Speech Extraction (TSE) in real-world conversational scenarios suffers from severe performance degradation due to the domain gap between synthetic training data and complex acoustic environments, where signal-level ground truth is typically unavailable. To address this challenge, we make the first attempt to extend RemixIT from
speech enhancement to TSE and propose a progressive synthetic-to-real adaptation framework for real-world TSE with two fine-tuning stages. The first stage leverages region-wise speaker similarity and silence constraints within a target-aware adaptation framework to jointly optimize the model using synthetic and weakly supervised real-world data, injecting real-world traits while preserving synthetic-learned capabilities. The second stage further adapts the model using only real-world data through our RemixIT-TSE, where quality-filtered teacher pseudo targets, which guarantee reliable student training, provide signal-level supervision via SI-SNR loss. Experiments on the real conversational evaluation set (EVAL-2) of the SLT 2026 REAL-TSE Challenge, the proposed method achieves a 6.53% relative TER reduction, together with relative improvements of 21.84% in speaker similarity, 9.89% in DNSMOS-P808, and 4.10% in target-activity F1 over the source-domain baseline, demonstrating its effectiveness under unseen real-world conditions. Source code at https: //github.com/YuWang-
Speech/RemixIT-TSE.
Read the full abstract ▾
All authors Yu Wang, Haixin Guan, Shuang Wei, Yanhua Long
Artificial Intelligence
Tool Mediation Alters Refusal Mechanisms in Large Language Models
Abel Rodríguez, Giuseppe Garofalo, Lieven Desmet et al. (+1 more)
Large
language models (LLMs) are increasingly deployed with access to external tools, yet harmful tool-mediated interactions are less likely to be refused when compared to regular conversational ones. As this change in refusal behavior remains underexplored, we investigate its underlying mechanisms across a diverse set of open-weight
language models. We find that information about the harmfulness of a request remains strongly encoded in the model's representations and transfers across conversational and tool-mediated inputs. Evidence from representation geometry and
neuron-level analysis further indicates that the two interaction modes systematically distribute harm-related computation differently. Crucially, while conversational inputs can be refused at relatively low levels of perceived harmfulness, tool-mediated inputs remain permissive until harmfulness crosses a substantially higher effective refusal threshold. Moreover, tool-mediated refusal is also more brittle: progressively weakening the refusal computation disrupts tool-mediated refusal at lower intervention strengths than conversational refusal, even when benign capabilities remain intact. Together, our findings indicate that tool mediation does not simply reduce the internal perception of harm, but instead impacts its conversion into refusal. Overall, this suggests tool-mediated environments may intrinsically reduce robustness of models to harmful requests, and that conventional safety evaluations may not fully transfer to LLM
agents.
Read the full abstract ▾
All authors Abel Rodríguez, Giuseppe Garofalo, Lieven Desmet, Vera Rimmer
Artificial Intelligence
DuplexCadence: Exact State and Execution from a Speech Model's Declared Timelines
Haixiao Gao, Yimin Zheng, Linyou Xiao et al. (+1 more)
Full-duplex
speech models support streaming interaction that listens and speaks at the same time. Serving them is governed by a strict, repeating deadline: conversation advances on a one-second cadence, and every second of input must be turned into a second of
speech before the next second arrives. Because stages within a session run in strict sequence, per-invocation overhead cannot be batched away. Profiling reveals that the autoregressive stages of a duplex second already fit within the period, whereas the
token-to-audio synthesis tail is what causes overruns. This tail stage suffers from orchestration slack where the GPU is left waiting as thousands of tiny, regular operations are issued one by one, while also wasting substantial
memory by over-provisioning state at static implementation constants. Existing remedies, such as graph recording and demand-sized allocation, fail because streaming state dynamics violate their prerequisites. The root cause is that the runtime lacks the model's native clocks: the per-region counters that govern advancement rates and retention policies. We propose DuplexCadence, which explicitly declares native clocks to the runtime and derives two mutually enabling rules: demand-sized state allocation at a stable address, and exact-shape graph replay without padding. The former eliminates idle
memory and stabilizes tensor pointers, while the latter removes orchestration slack without padding overhead. Evaluated on four released models across three decoder architectures with bit-for-bit identical output, DuplexCadence reaches $2.85\times$ the stock runtime's speed at $38.8\%$ lower peak
memory. On the live duplex path, mean SPEAK time falls from $14\%$ over the one-second cadence to $2\%$ under it, enabling models to reliably keep up with interactive
speech while markedly expanding multi-
Read the full abstract ▾
All authors Haixiao Gao, Yimin Zheng, Linyou Xiao, Zeke Xie
Machine Learning
SymbolicArena: A Unified Infrastructure for Benchmark Distillation and Dynamic Evaluation in Symbolic Regression
Ziwen Zhang, Xiju Wu, Yuheng Jing et al. (+9 more)
Symbolic regression (SR) seeks concise and interpretable mathematical expressions from data for scientific equation discovery. Existing SR
benchmarks face a tradeoff between evaluation cost and
benchmark validity. Repeated evaluation of large task pools is expensive, and compact
benchmarks lack systematic evidence of preserved task diversity and
algorithm discriminability. SymbolicArena provides a unified infrastructure for
benchmark distillation and dynamic evaluation. The framework standardizes 664 heterogeneous tasks with executable ground truth expressions and distills the Full Task Set into Core50, a validated
benchmark of 50 tasks. The
distillation process preserves task coverage and
algorithm discrimination under explicit balance constraints. SymbolicArena applies a unified execution protocol to heterogeneous SR
algorithms and produces comparable outputs and search trajectories. Multi Axis Evaluation characterizes numerical quality, symbolic quality, and search behavior. Core50 reduces evaluation workload by 92.5% and maintains agreement with Full Task Set evaluations. Experiments show that SymbolicArena achieves 72.6% to 86.7% lower approximation error than alternative selectors, further supporting its fidelity to the Full Task Set. Evaluation reveals a substantial gap between numerical fitting and symbolic recovery across current SR methods, suggesting that reliable equation recovery remains an open challenge.
Read the full abstract ▾
All authors Ziwen Zhang, Xiju Wu, Yuheng Jing, Runxiang Wang, Boxiao Wang, Yifan Zang, Yifan Zhang, Yang Wang, Kai Li, Yifan Zhang, Huilin Xu, Jian Cheng
Artificial Intelligence
Sol-H3: Recursive Self-Improvement for MiniMax-H3 Inference Acceleration on Sol-Engine across Cloud and Edge
Yitong Li, Jincheng Yu, Junsong Chen et al. (+6 more)
Video diffusion models are rapidly scaling and exhibiting enhanced generation capabilities. Among these recent advancements, MiniMax-H3 stands out as a highly capable, production-level open-source model. However, its 33-billion parameters and multi-step iterative denoising process introduce substantial computational overhead. Consequently, their practical production is hindered by generation latency in the cloud deployment like NVIDIA-GB200, alongside strict
memory limits that pose further challenges at the edge device like DGX-Spark. To address these diverse hardware bottlenecks from cloud to edge device, we present a full-stack inference pipeline that integrates efficient algorithmic design with optimized operator implementations. Algorithmically, we introduce a cross-resolution two-stage generation scheduler that exploits the step-wise nature of
diffusion: early low-resolution steps rapidly establish the global layout, while later high-resolution steps focus refinements of local and perceptual details. These stages are connected by a learned latent-to-latent mapping module, completely eliminating the computationally expensive VAE decode-reencode cycle for resolution transferring cross different resolutions. For operator implementation, we deploy a Recursive Self-Improvement (RSI) loop that searches kernel fusions and
memory layouts, evaluating latency together with numerical agreement. Together, these optimizations deliver up to 30x end-to-end speedup and 20% lower
memory: a 5-second 1344x768
video with audio is generated 3.5x faster than real time on an 8xGB200 node, and in under a minute fully
memory-resident on a single DGX Spark.
Read the full abstract ▾
All authors Yitong Li, Jincheng Yu, Junsong Chen, Haopeng Li, Shuchen Xue, Haozhe Liu, Ping Luo, Song Han, Enze Xie
Artificial Intelligence
Using Context Is Not Enough: Test-Time Training for Personalized Reward Modeling
Bohao Wang, Xiaoyan Zhao, Yang Zhang et al. (+4 more)
Reinforcement learning from human feedback (RLHF) aligns large
language models (LLMs) with human preferences, yet most pipelines learn a single reward model that overlooks individual differences in preferences. Personalized reward models (PRMs) address this by conditioning rewards on user-specific feedback, most commonly through in-context learning (ICL), where a user's historical comparisons are supplied as contextual preference pairs. However, we identify a key limitation of ICL-based PRMs: they fail to capture the preference relations conveyed by contextual pairs. To address this, we propose Preference-Aligned Test-Time Training (P-TTT), which explicitly encodes these relations into user-specific fast weights for personalized reward prediction. P-TTT introduces sequence-level update and apply operations to match the response-level granularity of preference feedback, together with a preference-aligned objective that directly uses pairwise preference relations to guide fast-weight adaptation. Notably, P-TTT is simple to implement and computationally efficient, updating fast weights within a single forward pass without inference-time backpropagation. Extensive experiments show that P-TTT more effectively captures historical preference relations and outperforms state-of-the-art methods by a large margin.
Read the full abstract ▾
All authors Bohao Wang, Xiaoyan Zhao, Yang Zhang, Jinghang Guo, Chun Chen, Can Wang, Jiawei Chen
Computation and Language
A mechanistic study of language model introspection
Jiahong Zou, Xiangkun Sun, Lingkai Kong et al. (+1 more)
Large
language models (LLMs) can sometimes report perturbations to their internal activations—even when the input provides no evidence that an intervention occurred. How do models detect and localize such internal changes? We study this question using a controlled task that keeps the input text fixed. We either inject a concept vector into the hidden state at one of ten
token positions or apply no intervention. The model is asked to identify the perturbed position or report that no intervention occurred. Across three model families, we identify two small groups of
attention heads with distinct roles in introspective reporting. Middle-layer gate heads influence whether the model reports a change, while router heads in a later layer help select the position to report. Interventions on gate heads can suppress position reports even when router heads supply location information. We further examine why reporting accuracy varies across concepts. Concept vectors that are localized more accurately produce stronger
attention-score and output responses in gate heads, which is associated with better
alignment of the induced key and value changes in their QK and OV computations. Together, these findings identify
attention-head mechanisms supporting introspective detection and localization.
Read the full abstract ▾
All authors Jiahong Zou, Xiangkun Sun, Lingkai Kong, Tonghan Wang
Artificial Intelligence
DoAtlas-2: A Foundation for Self-Evolving Causal Biomedical Discovery
Yulong Li, Rong Xia, Yuxuan Zhang et al. (+12 more)
We introduce DoAtlas-2, a foundation for self-evolving causal biomedical discovery that organizes knowledge around causal mechanisms and advances through external evidence from human populations. DoAtlas-2 integrates 771 research resources covering more than 720,000 participants in 48 countries, from longitudinal clinical phenotypes, medical imaging, and continuous physiological signals to eight molecular layers, together with an evidence network of approximately 4.7 million literature-derived records over 93,566 concepts and 149,383 candidate causal relations. DoAtlas-2 autonomously formulates research questions from evidence gaps and unresolved mechanisms, prespecifies their causal designs, and generates validated analyses. Supporting, challenging, and unresolved results continuously revise mechanistic interpretations, the causal evidence state, and the discovery frontier, so that DoAtlas-2 self-evolves within a closed loop of hypothesis generation, empirical testing, and renewed discovery. DoAtlas-2 has systematically evaluated 2,031 research questions. In the Human Phenotype Project (HPP), it formulated 4,014 candidate pathway questions across vascular, early-glycemic, and hepatic-metabolic systems, and screening of the first 1,079 yielded statistical support for 756. Representative studies identify blood pressure as a convergence node linking adiposity, hepatic, and lipid phenotypes to vascular outcomes, and show that an adiposity-inflammation-blood-pressure pathway is largely attenuated by joint adjustment for body mass index (BMI) and smoking. The discovered vascular network constitutes a completely interpretable predictive foundation, admitting exact attribution of every prediction and closed-form mediation effects. DoAtlas-2 thereby unifies causal mechanism discovery, population-evidence testing, and interpretable prediction within one continuously evolving foundation.
Read the full abstract ▾
All authors Yulong Li, Rong Xia, Yuxuan Zhang, Jianxu Chen, Xiwei Liu, Haochen Xue, Maosheng Li, Yuhang Liu, Yibo Yuan, Yutong Xie, Chong Li, Jionglong Su, Hagai Rossman, Eran Segal, Imran Razzak
Machine Learning
DRIFT: Disentangled Responsive-Invariant Flow Transport for Single-Cell Perturbation Prediction
Mustapha Bounoua, Giulio Franzese and Pietro Michiardi
Predicting cellular responses to perturbations is a central problem in cellular biology, with broad applications in systems biology and drug discovery. This task is challenging because cellular responses can be complex and
cell-state dependent, intrinsic
cell-to-
cell variability can be confounded with perturbation effects, and destructive single-
cell RNA sequencing precludes paired measurements of the same
cell before and after treatment. Flow matching transports control
cells to perturbed states flexibly, but acting on the full
cell state can confound perturbation effects with pre-existing
cell-to-
cell variability. Disentangled approaches separate responsive from invariant components, but model perturbations through prescribed mechanisms, such as latent shifts or graph edits, limiting their flexibility. We address both limitations in a unified framework. A variational encoder disentangles each
cell into an invariant block, capturing state unaffected by the perturbation, and a responsive block, capturing state it changes, through conditional priors and an information-theoretic invariance constraint. Conditional flow matching transports only the responsive block, conditioned on the perturbation and invariant state, yielding a flexible, data-driven model of perturbation effects without confounding pre-existing variability. Across several
benchmarks, our method outperforms the strongest published method in settings involving combinatorial and unseen perturbation prediction.
Read the full abstract ▾