Machine Learning
Human and LLM Facilitation in Collaborative Design Meetings: A Computational Behavioral Comparison
Mohammed Azeez Khan, Aaron Dsouza, Ashutosh Mishra et al. (+3 more)
Large
language models (LLMs) are increasingly being explored as facilitators of group discussion and collaborative work, but the behavioral differences between LLM-generated and human facilitation remain insufficiently characterized. We compare human Project Manager facilitation turns from scenario-based collaborative design meetings with Llama 3.1 8B Instruct responses across 199 paired contexts from the AMI Meeting Corpus. The analysis covers five confirmatory dimensions: contextual semantic distance, a lexical elaboration proxy, empathyand-hedge lexical density, topic divergence, and phase-vocabulary
alignment. An exploratory directiveness measure failed domain validation and is not interpreted. The sampled human Project Manager turns were higher on contextual semantic distance (r = 0.644) and on empathyand-hedge lexical density (r = 0.203). The second contrast is carried mainly by hedge terms, so we read it as lexical hedging rather than psychological empathy. Llama-generated facilitation was higher on topic divergence (r = 0.703), the lexical elaboration proxy (r = 0.288; lengthconfounded), and instruction-conditioned phase-vocabulary
alignment (r = 0.382). A secondary no-phase check left the Llama-higher phase contrast in place. The system
prompt still requested a facilitation move, so that check is not evidence of spontaneous phase awareness. These differences should not be interpreted as a ranking of facilitation quality. The resulting behavioral profile provides a reference point for later comparisons of LLM facilitators, human facilitators,
prompts, models, and collaborative settings.
Read the full abstract ▾
All authors Mohammed Azeez Khan, Aaron Dsouza, Ashutosh Mishra, Ummul Faiz Z. Bibi, Neha K. Nair, Arshaan Khan
Artificial Intelligence
SWEAR: Section-Wise Efficient Audio Reasoning via Self-Rated Difficulty
Yixuan Li, Minghao Lv, Xuenan Xu et al. (+2 more)
Recent work on large audio-
language models (LALMs) explores adaptive
reasoning to balance accuracy and efficiency. However, existing approaches often rely on coarse-grained difficulty estimation, overlooking the diverse factors that determine
reasoning effort. We address this limitation by deriving five-dimensional difficulty scores and compressed
reasoning traces from complementary offline assessments. Based on this characterization, we propose SWEAR, a difficulty-aware
reasoning framework that learns difficulty signals during supervised fine-tuning and uses reinforcement learning to adapt
reasoning allocation across
reasoning stages. Experiments on MMAU-test-mini and MMAR show that SWEAR improves accuracy while substantially reducing generated
tokens, shortening
reasoning for easy instances without sacrificing performance on difficult ones.
Read the full abstract ▾
All authors Yixuan Li, Minghao Lv, Xuenan Xu, Zihao Zheng, Mengyue Wu
Classical Analysis and ODEs
Limit cycles of slow-fast quintic Lienard systems
Hebai Chen, Leshan Chen and Yu Xiao
For every fixed real quintic
polynomial $F$, we prove that the slow–fast Liénard system $\dot{x}=y-F(x),\quad \dot{y}=-\varepsilon x$ admits at most two limit cycles whenever $0<\varepsilon<\varepsilon_0(F)$ for a suitable $\varepsilon_0(F)>0$. Using scalar comparison near the
equilibrium and a logarithmic fold-passage correction, we exclude \rev{canard singular limits}. Every sequence of limit cycles with $\varepsilon\to0^+$ has a subsequence converging in the Hausdorff metric to a common slow–fast path, whose regular slow arcs are either all normally attracting or all normally repelling. We also establish uniform Floquet separation for fixed $F$. The global counting argument uses shared regular arcs, nesting, and stability alternation. These results give a global small-parameter bound and describe the singular-limit structure, complementing the local cyclicity analysis of Li and Lu(J. Differential Equations, 2014).
Read the full abstract ▾
Robotics
VICAR: Video Imitation with Contact-Aware Retargeting for Humanoid Interactive Skills
Dvij Kalaria, Bike Zhang and S. Shankar Sastry
Third-person human
videos offer a scalable alternative to task-specific reward engineering, teleoperation, and manually scripted demonstrations for
humanoid loco-manipulation. Recent
video-to-
humanoid systems have shown that noisy human motion, object trajectories, and contacts can be extracted from unstructured
videos and refined into deployable skills. However, a key gap remains for contact-rich imitation: the task is often defined by world-frame contacts whose poses, timing, and speed must survive embodiment mismatch and runtime scene changes. Towards this, we present VICAR (
Video Imitation with Contact-Aware Retargeting), a pipeline for learning
humanoid interactive skills from third-person human
videos. The system first recovers human motion, reconstructs task-relevant obstacles from
monocular video, and then retargets the motion to a
humanoid while preserving global contact poses, motion speed, and collision clearance. Unlike retargeting methods that only remove penetration or preserve local interaction geometry, our formulation represents user-specified contact poses and collision clearance as high-priority Lagrangian penalty terms and optimizes the full motion to remain smooth through contact-rich transitions. The resulting motions are used to train a contact-conditioned motion generator whose inputs include movable contact points enabling runtime adaptation to new contact positions. We demonstrate the framework on ten diverse third-person
video skills, including six table-tennis serves, three object-pickup tasks, and ladder climbing, and deploy the learned skills on hardware. These results suggest that contact-preserving retargeting is a practical step toward
humanoids that learn useful interactive behavior by watching humans.
Read the full abstract ▾
Artificial Intelligence
Attention Sees, MLP Refuses: Tracing the Attention-to-MLP Safety Pathway in LLMs
Zixuan Weng, Haoqi He, Weifei Jin et al. (+2 more)
Safety-aligned large
language models (LLMs) remain vulnerable to jailbreaks and can over-refuse benign requests. Understanding these failures requires distinguishing harmfulness recognition from the computation that writes refusal. We reframe this as a pathway question and make it measurable on residual-stream updates in decoder-only
Transformers, separating the harmfulness information a module update exposes from the change it produces in readable refusal evidence. Across 18 Base-Instruct pairs spanning Qwen, Llama, and Gemma, harmfulness is decoded more readily from
attention updates, particularly in early layers, whereas MLP updates contribute more strongly to harm-selective refusal writing in aligned models. Post-training strengthens MLP refusal writing in all 18 pairs without systematically improving harmfulness decodability, while positive MLP writing becomes more concentrated in deeper layers after post-training. Across six
reasoning-enabled models, harmfulness remains decodable during
reasoning while strong refusal writing concentrates at answer onset, a pattern that persists with
reasoning disabled. Causal interventions further support this pathway: resetting downstream MLP updates to their clean values substantially reduces the refusal shift caused by upstream
attention interventions. Building on these findings, we introduce Sentinel, which combines an
attention-derived perception score P with an MLP-derived decision score D. Their joint profiles characterize jailbreaks and over-refusals, and conditional routing with both signals achieves a lower aggregate error rate than single-signal routing and evaluated fixed reminders across eight models.
Read the full abstract ▾
All authors Zixuan Weng, Haoqi He, Weifei Jin, Yue Huang, Yang Bai
Robotics
Towards Safer Cities: A Design Process for Perception Datasets in Urban Driving Scenarios
Alexander Kroth, Cristian Martinez and Steven Peters
Urban environments present significant challenges for the perception systems of automated driving due to their high complexity, characterized by dense traffic, diverse obstacles, or a wide range of possible scenarios. While numerous automotive
datasets exist, few focus on urban settings, and the majority are acquired exclusively on passenger cars and overlook other transport modalities such as heavy-duty trucks, busses and trams. In contrast to passenger vehicles, these platforms operate under different geographical and operational constraints; including traversing pedestrian zones or platform-specific tasks such as movement of trailer or winter service. In this work a method is proposed for a guided
dataset creation based on an analysis of existing
datasets using user-generated map data. Furthermore, representative truck-specific use cases are identified and exemplified to demonstrate the applicability of the proposed methodology and to support targeted acquisition planning.
Read the full abstract ▾
Group Theory
Classes of Subgroups of a finite group with isomorphic quotient groups
Tushar Halder and Sachin Ballal
In this paper, we study the poset QIso(G) of classes of subgroups of a finite group G having isomorphic quotients. Particularly, we study it for finite abelian groups and for some classes of non-abelian metacyclic groups, namely dihedral groups Dn, the group U2np and affine groups Aff(F) over finite field F. In the case of abelian groups G, we show that there exists a duality from QIso(G) to Iso(G). We show that cyclic groups are the only groups for which QIso(G), up to duality is isomorphic to its subgroup
lattice. We also show that cyclic groups are the only nilpotent groups for which QIso(G), up to duality is isomorphic to its
lattice of normal subgroups. We characterize finite 2-groups all of whose subgroups are q-self dual with QIso(G) being a chain. We also characterize all p-groups G of order at most p4 whose QIso(G) is a chain.
Read the full abstract ▾
Artificial Intelligence
LongHarness: Advancing Agentic Reasoning over 100M-Token Corpora via Stateful Evidence Management
Songtao Huang, Zhijie Zhong, Xiaohan He et al. (+8 more)
Corpus-level long-context
reasoning requires
agents to acquire and integrate evidence across document collections that exceed their context windows. Existing
agent harnesses enable incremental access to these corpora, but acquired evidence and evolving conclusions often remain entangled in a growing interaction history. This implicit state management can lead to two coupled failure modes: unbounded exploration, where
retrieval expands the interaction history without resolving decision-relevant information gaps, and evidence-state drift, where previously acquired evidence or subsequent corrections fail to inform the current conclusion. We present LongHarness, a model-agnostic harness that makes the interaction between acquired evidence and evolving conclusions explicit. Its core mechanism, Evidence-Conclusion Coevolution, uses unresolved conclusions to guide exploration toward missing information, while newly acquired evidence strengthens or revises the conclusions that directed the search. LongHarness couples this process with Evidence Lifecycle Management, which keeps accumulated evidence current and traceable throughout the interaction. Together, these mechanisms focus
retrieval on remaining information needs and keep conclusions aligned with current evidence. We evaluate LongHarness on OfficeQA Pro V1 and V2, AA-LCR, and LongBench v2 using GLM-5.3-Flash and Qwen3.8-Flash. Compared with direct
reasoning using the same backbone, LongHarness improves GLM-5.3-Flash exact match on OfficeQA Pro from 15.8% to 60.2% on V1 and from 4.4% to 61.1% on V2. Against state-of-the-art
agent harnesses, it surpasses Claude Code by 12.0 percentage points on V1 and Codex by 18.9 points on V2. With Qwen3.8-Flash, LongHarness reaches 61.7% on V1, outperforming Claude Code by 8.3 points and Codex by 12.0 points.
Read the full abstract ▾
All authors Songtao Huang, Zhijie Zhong, Xiaohan He, Shuaiyu Zhang, Zhiyin Yu, Li Kang, Heng Zhou, Yutao Fan, Xiangchao Yan, Bo Zhang, Shiyang Feng
Quantum Physics
All-optical deconvolution via quantum cross-correlation spectroscopy with bright twin beams
Yujiro Eto
The resolution of spectroscopic measurements is fundamentally limited by the convolution between a system's intrinsic response and the instrument response function, creating an ill-posed inverse problem. While entangled
photons can theoretically bypass these limits, low
photon flux has hindered their experimental realization. Here, we propose
quantum cross-correlation spectroscopy (QCCS), utilizing macroscopic twin beams (signal and idler beams) to physically circumvent this convolution limit in coherent anti-Stokes Raman scattering. Employing the signal beam as a broadband probe, QCCS achieves all-optical deconvolution through a correlation-enhanced sum-frequency generation channel between the convoluted anti-Stokes and idler beams. By subtracting the uncorrelated background using a 500-femtosecond temporal window, we successfully retrieve narrow Raman bands (8 cm^{-1} and 15 cm^{-1}) completely obscured by a 75 cm^{-1} probe bandwidth without numerical deconvolution or prior knowledge. Our results establish macroscopic
quantum correlations as a robust resource for high-fidelity spectroscopy, paving the way for ultrashort-pulse-free operando studies under natural, incoherent excitation conditions.
Read the full abstract ▾
Machine Learning
Double-Take: Improving Capacity, Binding, and Selectivity In Long-Context Within Bidirectional Memory
Vakul Raghavan, Utkarsh Grover, Mingyang Mao et al. (+2 more)
Processing a long sequence does not guarantee that distant information remains usable. As context grows, a fixed-size
memory must retain many associations, connect information across positions, and preserve relevant structure without being overwhelmed by everything it observes. These are distinct requirements: a
memory may store more information without representing the relationships needed for
retrieval, and it may transport those relationships without preserving the ones that matter. Thus, the challenge is not simply extending context length, but constructing a
memory that remains usable as context grows. We study this problem in large-chunk test-time training (TTT) and identify two basic failures: single-step writes accumulate interference, limiting capacity, while self-associative writes cannot form cross-position bindings. We introduce Double-Take (DT), which repairs these failures with multi-step unpinned updates and cross-position mixing, and adds bidirectional change-sensitive writes for selective retention. Capacity alone leaves end-to-end recall at the $0.10$ chance floor; adding binding raises it to $0.92$. Trained only on $32\,\mathrm{kb}$ genomic contexts, DT reaches $0.927$ AUROC at $131\,\mathrm{kb}$ and $0.856$ near $1\,\mathrm{Mb}$, while transferring zero-shot from human to mouse. Together, these results identify capacity, binding, and selectivity as three requirements for usable long-range
memory.
Read the full abstract ▾
All authors Vakul Raghavan, Utkarsh Grover, Mingyang Mao, Chengwen Wang, Xiaomin Lin
Sound
AV-FlowEdit: Training-Free Prompt-Driven Audio-Visual Editing with Event-Aware Spectral Trajectory Preservation
Shaohua Pan and Xiao Liu
Joint audio-
video diffusion models generate synchronized content, but editing existing clips requires target changes without disrupting non-target structure or cross-modal correspondence. We present AV-FlowEdit, a training-free and inversion-free framework for
prompt-driven audio-
video editing. Our key insight is that source–target flow differences provide both cross-modal editing directions and preservation cues. Under shared perturbations, they jointly update the
video and audio branches, supporting visual-only, audio-only, and joint edits while retaining pretrained cross-modal interactions. To suppress drift accumulated during sampling, Cross-Modal Event-Aware Spectral Trajectory Preservation combines flow saliency, aligned audio activity, multiscale low-frequency displacement, and
diffusion-stage scheduling. This adaptively preserves unsupported regions while allowing intended audio-visual events. AV-FlowEdit requires no model training, external masks, or optimization-based inversion. Experiments show strong
prompt alignment and temporal consistency, together with competitive audio quality and cross-modal coherence.
Read the full abstract ▾
Cryptography and Security
CAPE - Context-Aware Prioritization Engine for Vulnerability Scoring and Management
Rakesh Podder, Viktoria Koscinski and Indrajit Ray
Vulnerability patching is critical for safeguarding an enterprise's software ecosystem from cyber attacks. However, patching requires adequate preparation for potential system downtime and for unforeseen effects of the patched software on a live system. Thus, effective risk mitigation requires prioritization strategies that explicitly account for the operational context of the affected software. While vulnerability prioritization based on Common Vulnerability Scoring System (CVSS), Exploit Prediction Scoring System (EPSS) and CISA Known Exploited Vulnerabilities (KEV) catalog, can help, their differing goals and methodologies often lead to inconsistent, context-unaware decisions resulting in mis-allocated remediation efforts. In this work, we present the Context-Aware Prioritization Engine (CAPE), a vulnerability analysis framework that enriches each vulnerability scanner report of vulnerabilities (CVEs) with deployment-specific evidence. This evidence is used to perform vulnerability reachability, centrality, and exploitability analysis, producing a ranked priority score for the effective distribution of remediation effort. We evaluate CAPE on 30 open-source agentic applications comprising 6,138 scanner-reported CVEs. Our evaluation shows that a significant majority of CVEs are not reachable and exploitable which results in significant disparities between CAPE and context-free baselines - the CVSS, EPSS, and KEV. It also shows that CAPE consistently surfaces confirmed-exploitable CVEs at higher priority than all baselines, demonstrating that context-aware analysis provides substantially more accurate remediation guidance than context-free severity scoring alone.
Read the full abstract ▾
Optimization and Control
A Palais-Smale type condition for optimal control problems
Jean-Michel Coron, Li Deng and Xu Zhang
We consider an optimal control problem with a general control constraint and a fixed initial state, for which convexity assumptions are absent. To establish the existence of optimal pairs, we introduce a control version of the Palais-Smale type condition adapted from critical point theory and prove that optimal pairs exist under this condition (see
Theorem 2.1). Subsequently, we verify that a specific class of optimal control problems satisfies our Palais-Smale type condition (see Proposition 2.1). Furthermore, as an application, we present an example (Example 3.1) and show that, to the best of our knowledge, no existing result can be applied to establish the existence of an optimal pair in this case (see Section 3.1.1). Example 3.2 shows that finitely many Pontryagin maximum principle candidates do not guarantee the existence of an optimal control. Finally, Example 3.3 demonstrates that the proposed Palais-Smale type condition is sufficient, but not necessary, for the existence of optimal controls.
Read the full abstract ▾
Probability
Penalizations of Two-Dimensional Simple Random Walk by Direction-Dependent Excursion Counts and Edge-Crossing Counts
Syota Esaki and Kohki Iba
We study penalizations of the two-dimensional simple random walk by functions of direction-dependent excursion counts and edge-crossing counts. The former record completed excursions with prescribed pairs of departure and return directions at the origin, whereas the latter record crossings of a prescribed set of unoriented edges incident to the origin. We determine the distributions of their almost surely finite terminal values and derive explicit path decomposition formulas involving the random walk conditioned to avoid the origin. Moreover, both families of penalized measures can be represented by weighting and normalizing the same universal $\sigma$-finite measure associated with local-time penalization at the origin. Thus, direction-dependent excursion counts, edge-crossing counts, and local time at the origin give rise to penalizations belonging to a common universality class.
Read the full abstract ▾
Dynamical Systems
Uniform Exponential Attractors for Non-Autonomous Dynamical Systems and an Application to the damped hyperbolic equation with sub-quintic growth
Chang Qingquan and Li Dandan
In this paper, we develop an abstract framework for the theory of uniform exponential attractors in non-autonomous dynamical systems. Our construction employs time-dependent absorbing sets while ensuring exponential attraction among the constructed sets. Moreover, we show that the resulting exponential attractors are uniform-that is, they remain stable under perturbations. As an application of this abstract theory, we study the weakly damped
hyperbolic equation with sub-quintic nonlinearity. Using the proposed approach, we construct exponential attractors for the process generated by the Shatah-Struwe solution, which satisfies both energy and Strichartz estimates. Furthermore, we establish the Holder continuity of these exponential attractors with respect to the external forcing term.
Read the full abstract ▾
Hardware Architecture
TPCS: A Cycle-Level Simulation Paradigm with Native Execution Semantics for RTL Modeling and Generation
Chengzhen Meng, Xiaoyu Li, Chen Bai et al. (+2 more)
The growing diversity of domain-specific architectures and microarchitectural designs demands fast and accurate evaluation methods during early design stages. However, existing approaches suffer from a fundamental trade-off: high-level simulators offer speed but lack fidelity, while RTL-based tools provide accuracy at the cost of significant simulation overhead. This gap limits efficient pre-RTL microarchitecture exploration. Therefore, we propose Transaction-Pruned Cycle Simulation (TPCS), a novel modeling and simulation framework. It serves as a unified intermediate abstraction spanning microarchitecture modeling, cycle-accurate simulation, and synthesizable RTL implementation. TPCS organizes combinational logic as transaction-based function invocations and leverages host CPU control flow to naturally prune inactive logic, eliminating the need for explicit event scheduling or signal tracking. This enables cycle-accurate simulation with significantly improved performance. TPCS preserves structural correspondence with RTL and supports automatic generation of synthesizable designs. TPCS achieves 36$\times$-42$\times$ speedups over parallel Verilator on SMP systems and 3.4$\times$-7.2$\times$ over single-threaded Verilator on accelerators, with cycle-level waveform agreement on the evaluated traces. The generated RTL demonstrates comparable timing and area to handcrafted implementations, with relative changes of -4.7% to +2.6% in clock period and -3.3% to +4.0% in standard-
cell area. By bridging the abstraction gap between high-level modeling and RTL design, TPCS enables a unified and agile pre-RTL workflow, allowing efficient microarchitecture exploration with both high performance and implementation fidelity.
Read the full abstract ▾
All authors Chengzhen Meng, Xiaoyu Li, Chen Bai, Zhenyu Zhang, Hongjun Dai
Computer Vision
Evolving Creatives with Clicks: CTR-Driven Advertising image Editing via Attribute Disentanglement and Preference Alignment
Xu Zhang, Haohan Wang, Zheng Zhang et al. (+9 more)
The optimization objective for e-commerce advertising creatives has shifted from aesthetic quality to online business metrics like click-through rate (CTR). Recent methods generate appealing ad images from product images by constructing CTR-based reward signals and refining
prompt generation via Reinforcement Learning. Such paradigms equally penalize all attributes within an underperforming image, even though some are actually beneficial to clicks. In this work, we propose Evolving Creatives with Clicks (ECC), which reformulates creative optimization as fine-grained editing rather than generation from scratch. ECC selectively refines attributes that hurt CTR while preserving those that contribute positively. Specifically, faced with a rich attribute space across massive images, ECC first identifies 28 attributes significantly correlated with CTR via ANOVA and Random Forest regression. Based on attribute contrasts between images with disparate CTRs, we instruction-tune a
prompt model to generate editing suggestions, then refine it with real click preferences since reference-derived suggestions are not uniformly beneficial. Furthermore, we introduce Consistent Preference Optimization (CPO) as ECC's preference optimization strategy, which enforces consistency across semantically equivalent shuffled suggestions to mitigate DPO overfitting and improve preference generalization. Extensive experiments demonstrate that ECC significantly improves the CTR of ad images, outperforming state-of-the-art models by a substantial margin. The complete data pipeline, including attribute extraction and analysis, the resulting
dataset, and source code will be released upon acceptance.
Read the full abstract ▾
All authors Xu Zhang, Haohan Wang, Zheng Zhang, Yingjie Xia, Wei Feng, Honghe Zhu, Jingjing Lv, Changping Peng, Xuheng Qin, Zehua Zhang, Zheng Luo, Qicheng Li
Software Engineering
TestSuiteBench: Coding Agents Struggle to Generate Strong Test Suites from Specifications
Kazuki Egashira, Niels Mündler-Sasahara, Mark Niklas Müller et al. (+1 more)
Coding
agents are increasingly used to create large software projects. While code-generation
benchmarks have scaled accordingly to full applications, test-generation
benchmarks remain limited to individual functions or bug fixes. In particular, generating complete test suites for complex software from scratch remains unexplored. To address this gap, we introduce TestSuiteBench, a
benchmark of 165 open-source applications, each with a structured and detailed specification of intended behavior, a reference implementation, a reference test suite, and on average 74 mutants. Each mutant is generated from a single requirement of the specification and refined to isolate that behavior: 73% are killed by exactly one reference test and, on average, by only 0.2% of the remaining tests. We find that even frontier coding
agents struggle to generate strong test suites given our specifications, achieving mutation scores below 60%, despite their suites being largely valid and 88% of their tests passing on the reference implementation. We observe that generated test suites often lack relevant input combinations and are smaller than reference test suites. Meanwhile, implementations generated for the same specifications pass 82% of the reference tests. These results show
agents can build software they cannot verify, a gap that directly limits fully autonomous software development.
Read the full abstract ▾
All authors Kazuki Egashira, Niels Mündler-Sasahara, Mark Niklas Müller, Martin Vechev
Machine Learning
CB-Orchestrator: Adaptive Workflow Optimization for LLM Agents via Contextual Bandits
Jiahang Sun, Zhiwei Shang, Zhipiao Liu et al. (+4 more)
Large
Language Model (LLM)
agents have demonstrated remarkable capabilities in tackling complex tasks through agentic workflows. However, manually designing these workflows is labor-intensive and lacks flexibility to adapt to diverse queries. While automated workflow optimization is a promising direction, existing methods often incur expensive API costs during training-phase feedback collection. Furthermore, they fail to adaptively reuse high-quality workflows, which creates a bottleneck for both performance and cost-effectiveness. To address these limitations, we propose CB-Orchestrator, an adaptive framework that decouples workflow generation from selection. In the first stage, we construct a diverse pool of candidate workflows via evolutionary search. In the second stage, we formulate workflow selection as a Contextual Bandit problem, which enables sample-efficient learning by balancing exploration and exploitation, thereby significantly reducing the feedback required for training. During inference, the model adaptively selects the optimal workflow tailored to each query. Evaluations across five diverse
benchmarks demonstrate that CB-Orchestrator consistently outperforms all baselines. Notably, on the MATH
dataset, CB-Orchestrator achieves better performance than strong recent baselines while reducing training-phase API
token overhead by no less than 40.55% and total end-to-end training (including workflow pool generation)
token overhead by no less than 29.73%.
Read the full abstract ▾
All authors Jiahang Sun, Zhiwei Shang, Zhipiao Liu, Hongwei Yang, Guoqing Xie, Shuang Qiu, Zhongxiang Dai
Probability
The Order Mass Identity of Rooted Forests, with a Direct Proof of the Extended Aldous-Broder Theorem
Hugh Zheng
The Aldous-Broder
algorithm constructs a spanning tree from the first-entrance edges of a covering random walk. Its distribution is known, but deriving it by summing over the orders in which a fixed tree can be discovered remains difficult. In this paper, we evaluate this sum by establishing an order-mass identity for rooted forests. For real
matrices with zero row sums, the identity expresses the sum of order weights associated with each root as a principal cofactor of a
matrix on the roots, under a nonsingularity condition involving only the sets encountered along the growth orders of the given forest. We prove the identity using linear algebraic tools. With one root, the sum equals one, giving a direct proof of the extended Aldous-Broder
theorem for every finite irreducible Markov chain started at a fixed state. With multiple roots, we derive the forest law produced by the stopped covering walk and show how it differs from the weight-proportional forest law. The identity and the resulting tree and forest laws are formalized in Lean 4.
Read the full abstract ▾
Astrophysics of Galaxies
A Relationship Between Fast Radio Burst Dispersion Measure and Host Galaxy Inclination
Lachlan Marnoch, Stuart D. Ryder, Sunil Simha et al. (+9 more)
It is commonly thought that fast
radio bursts (FRBs) trace
star formation, and that they are produced by magnetars. Both hypotheses imply that their sources are hosted primarily (but not exclusively) in
galactic disks. Although there is evidence in that direction, it has yet to be shown conclusively. In addition, a tension exists with the fact that at least one FRB source is found in a globular cluster far from the disk of the host. If FRBs originate primarily in disks, the
galaxy orientation could leave an impression in the sample of observed FRBs, either as a selection bias against FRBs in edge-on
galaxies or as a correlation with DMHost. Statistical tests against the inclinations of FRB host
galaxies, derived from parametric morphology applied to optical and near-infrared imaging, can search for these effects. Such a search requires an adequately large sample of known FRB hosts that have been imaged to a sufficient depth, while estimates of DMHost also require the
redshift of each host to be known. These conditions are best met in the Commensal Real-time ASKAP Fast Transient (CRAFT) survey sample of FRB host
galaxies, for which the majority have been imaged at optical wavelengths with FORS2 on the Very Large
Telescope (VLT). Here these data are fitted with the morphology software GALFIT to derive the elliptical axis ratio q (a proxy for inclination) for 34 hosts from the CRAFT sample, detected with ASKAP and followed up with VLT imaging. We find no bias against FRB detections from edge-on
galaxies; instead, we uncover a new relationship between the inclination of an FRB host
galaxy and DMHost. This is consistent with a primarily disk-based origin for CRAFT FRBs. Using this relationship we are able to reduce the scatter of the host sample about the Macquart relation by 40%, demonstrating a new empirical method for estimating DMHost that requires only deep optical imaging.
Read the full abstract ▾
All authors Lachlan Marnoch, Stuart D. Ryder, Sunil Simha, Alexa C. Gordon, R. M. Shannon, Nicolas Tejos, Marcin Glowacki, Ziteng Wang, Elizabeth K. Mahony, J. Xavier Prochaska, Wen-fai Fong, Richard M. McDermid
Distributed Computing
Capturing Electrical Windows Through Reconfigurable Optical Networks: Transfer-aware Scheduling of Multi-Region AI Data Centers
Xi Tong, Zijian Li, Xiaoyi Fan et al. (+3 more)
Artificial-intelligence data centers (AIDCs) can provide spatiotemporal load flexibility by migrating training workloads in response to regional electricity prices and renewable-energy (RE) availability. However, transfer-blind scheduling may overestimate migration benefits by overlooking transfer time or prevent training from resuming within renewable-rich or low-price windows when considering transfer time. This paper proposes a transfer-aware compute-electricity-network scheduling framework that coordinates training placement, checkpoint transmission, and optical
circuit switching (OCS) to improve local RE utilization and reduce grid-purchase cost. A compact compute-electricity master problem generates candidate schedules using estimated transfer delays. The framework then schedules optical
circuits, verifies checkpoint delivery under shared-link capacity constraints, and revises infeasible schedules using updated delay estimates. It integrates day-ahead planning, event-triggered intraday updates, and real-time dispatch. In a five-AIDC
benchmark with equal completed work and network resources, the proposed method reduces the real-time grid purchase cost by 12.50% and increases the local RE utilization by 2.724 percentage points relative to static optical operation. Sensitivity and ablation analyses indicate that OCS is particularly beneficial when renewable-rich or low-price windows shift across sites and network bottlenecks hinder timely checkpoint delivery.
Read the full abstract ▾
All authors Xi Tong, Zijian Li, Xiaoyi Fan, Cong Zhang, Ershun Du, Qin Wang
Computation and Language
KV-Repurpose: A Testbed for Evaluating Non-Prefix KV-Cache Reuse
Taiga Ikeda, Daisuke Miyashita and Jun Deguchi
KV-cache reuse in large
language model serving is commonly limited to exact
prompt prefixes, preventing reuse when the same retrieved document appears after different system
prompts, conversation histories, or dynamic metadata. We present KV-Repurpose, an open-source, vLLM-based system for developing and evaluating segment-wise KV-cache reuse beyond exact prefixes. The system combines cache-aware segment scheduling,
prompt construction, capacity management, workload generation, and runtime monitoring. Experiments on long-context
benchmarks show that naive segment reuse largely preserves output quality for single-document question answering and summarization, while exposing clear limitations in cross-document
reasoning. Under a Zipfian RAG-style workload in which 20% of requests share a fixed prefix, KV-Repurpose achieves a substantially higher
token-level cache hit rate than prefix caching at the same 30 GB cache capacity, 63.9% versus 3.3%, resulting in lower TTFT. Even with 1 TB of cache capacity, prefix caching reaches only a 17.9% hit rate in this workload. Source code is available under the Apache-2.0 license at https://github.com/taiga333/eval-kv-repurpose.
Read the full abstract ▾
Computation and Language
CliQ-RRG: Clinical-Knowledge Guided Disease-aware Visual-Textual Alignment for QA-Style Radiology Report Generation
Chandravardhan Singh Raghaw, Shahid Shafi Dar, Arnav Nishith Sharma et al. (+2 more)
Radiology reports are essential for diagnostic
reasoning and patient care, yet their manual preparation is time-consuming and cognitively demanding. Automatic radiology report generation (RRG) offers a scalable alternative, but existing models often produce lengthy, unstructured narratives that overlook diagnostic cues and multi-view information. We present Clinical-Knowledge Guided Disease-aware Visual-Textual
Alignment for QA-Style Radiology Report Generation (CliQ-RRG), a unified two-stage framework for interpretable and clinically grounded reporting. In Stage 1, CliQ-RRG employs a Disease-aware Visual-Textual
Alignment module that aligns image and text representations using predicted disease
embeddings, reinforced by a Prior-Guided
Attention Module (PrAM) to capture multi-view dependencies across current and prior scans. In Stage 2, GraphRAG-guided
retrieval augments intermediate textual representations with graph-grounded clinical evidence, and a large
language model restructures them into concise, interpretable question-answer (QA) pairs with diagnostic summaries. Experiments on two public chest X-ray
benchmarks demonstrate that CliQ-RRG consistently outperforms prior methods across both lexical and clinical metrics, generating accurate and clinically coherent QA-style radiology reports. Code is available at: https://github.com/chandravardhan-singh-raghaw/CliQ-RRG.
Read the full abstract ▾
All authors Chandravardhan Singh Raghaw, Shahid Shafi Dar, Arnav Nishith Sharma, Shrey Modi, Nagendra Kumar
Artificial Intelligence
Travel Back To Move Forward: Retrospective Exploration and Evolving for Long Horizon Agents
Ziang Ye, Wentao Shi, Lei Chen et al. (+7 more)
Large
language model
agents increasingly operate in long-horizon environments where early mistakes can trigger cascading failures and trap
agents in dead-end trajectories.
Agents must therefore be able to proactively recover from such states and turn their exploratory experience into guidance for subsequent decisions. We introduce ReTRAIL, a framework that enables
agents to rewind the environment to earlier states while preserving valid progress and carrying forward knowledge learned from abandoned exploration. The environment state moves backward while the
agent's knowledge continues to move forward. ReTRAIL trains a shared policy with two complementary signals: State-Anchored GRPO compares alternative continuations from revisited states to improve execution, while a branch-advancement reward guides reflection through the progress of subsequent exploration. Experiments across four interactive
benchmark MineSweeper, Sokoban, ALFWorld and ScienceWorld show that simply adding a rewind primitive provides limited benefit and can even degrade performance. In contrast, ReTRAIL enables
agents to rewind from unproductive states and use insights from abandoned branches to guide subsequent exploration. These results demonstrate the value of coupling environment rewinding with experience reuse within an ongoing task for long-horizon task solving. The code is available at https://github.com/whi497/ReTRAIL.
Read the full abstract ▾
All authors Ziang Ye, Wentao Shi, Lei Chen, Zhengzhou Cai, Yu Wang, Zhiyuan Yao, Zihan Wang, Xierui Song, Fuli Feng, Yixin Cao
Computation and Language
CoBe: A Benchmark For Conversational Counterfactual Text Editing
Arvind Raghavan, Inwoo Hwang, Shreyas Havaldar et al. (+6 more)
Counterfactual text editing is the ability to revise a given text in light of a hypothetical change while preserving facts not causally affected by that change. When asked what would have happened if a doctor had not recommended surgery, the model output should retain the patient's medical history and modify only the consequences of the decision. This requires causal understanding: a model must determine, without being explicitly instructed to, what events are affected by the hypothetical change and what should remain fixed. This skill is central to human
reasoning, especially in safety-critical domains like healthcare or law. By contrast, the task of associational text generation merely generates text statistically associated with the altered condition. In the medical example, an associational response might make the patient's symptoms less severe because milder cases correlate with no surgery. Such responses produce a plausible scenario, but not the counterfactual version of the given one. Prior counterfactual-text
datasets do not clearly separate these modes of
reasoning. We introduce the Counterfactual
Benchmark (CoBe), a testbed for models' zero-shot ability to counterfactually edit a given textual scenario when prompted to do so in a conversationally natural way. CoBe contains 2000 questions spanning healthcare, science and engineering, business, and everyday scenarios. We test a suite of frontier and smaller models, with frontier systems achieving an average score of $58.05\%$. Our findings indicate that models frequently default to associational generation rather than preserving the causal structure of the original scenario. Moreover, performance does not scale consistently with either model size or output
token length, suggesting that the failure may reflect model training objectives rather than insufficient model capacity.
Read the full abstract ▾
All authors Arvind Raghavan, Inwoo Hwang, Shreyas Havaldar, Kai-Zhan Lee, Aurghya Maiti, Yushu Pan, Jeffrey Wu, Mingxuan Li, Elias Bareinboim
Computer Vision
A4-MR: Abnormality-aware Anatomical Alignment for MR
Noel D'souza, Zhijian Yang, Nishank Singla et al. (+14 more)
Recent advances in large-scale foundation models have enabled strong generalization across a wide range of clinical and research tasks in
magnetic resonance imaging (MRI). However, existing
vision-
language MRI models remain limited in anatomy-specific localization and pathology-aware representation, particularly in real-world settings. We introduce A4-MR, a 3D
vision-
language foundation model trained on approximately 50,000 MRI studies comprising 491,000 series, roughly half of which are paired with radiology reports. A4-MR is designed to: (1) learn anatomy localization through large-scale pretraining with noisy labels; (2) align image and text representations at the anatomical level; and (3) learn distinct representations of healthy and abnormal anatomy. Across a diverse set of downstream tasks, A4-MR outperforms existing foundation models, including a +1.7 AUROC improvement over MedSigLIP for abnormality detection. Using a frozen encoder, A4-MR also surpasses an end-to-end task-specific MedNeXt model for knee
segmentation by +1.4 Dice and performs comparably to the best task-specific model for abdominal organ
segmentation on AMOS, within 0.4 Dice. Finally, without anatomical
prompts, A4-MR localizes 3D anatomy comparably to MedSAM2 provided with ground-truth bounding boxes. Together, these results demonstrate that anatomy-level
multimodal alignment and abnormality-aware representation learning can enable a single MRI foundation model to support localization,
segmentation, and pathology-related tasks.
Read the full abstract ▾
All authors Noel D'souza, Zhijian Yang, Nishank Singla, Pranay Doshi, Krisztian Koos, Fei Tan, Bruno Nunes, Laszlo Rusko, Bernadett Kolozsvari, Eszter Sabo, Emanuele Valeriano, Vikash Gupta, Seyed Iman Zare Estakhraji, Marc Lebel, Farzin Haddadpour, Taha Kass-Hout, Erhan Bas
Group Theory
Cohomology and multiplicative gaps for quasi-isometries of nilpotent groups
Mayukh Mukherjee, Soumyadeb Samanta and Soumyadip Thandar
We construct two word metrics on the same seven-dimensional torsion-free nilpotent
lattice whose asymptotic cones are isometric, but for which every quasi-isometry has multiplicative constant at least $1+\varepsilon$, for one fixed $\varepsilon>0$. We use a convex invariant of a fixed quasi-isometry $\Psi:N\to M$. The displacement cocycle of $\Psi$ and the right-invariant means on $N$ determine a compact convex set $\mathcal A(\Psi)\subseteq\operatorname{Hom}(V_N,V_M)$. If $\mathscr G_{\mathrm{pt}}(\Psi)$ denotes the first-layer maps of the graded homomorphisms that occur as blow-downs of $\Psi$ with varying basepoints, then $$ \mathcal A(\Psi)=\operatorname{conv}\mathscr G_{\mathrm{pt}}(\Psi), \qquad\operatorname{Ext}\mathcal A(\Psi)\subseteq\mathscr G_{\mathrm{pt}}(\Psi). $$ For every extreme point $L\in\mathcal A(\Psi)$, the dual map $L^*$ is also the degree-one part of an isomorphism of the full real
cohomology rings. This cohomological compatibility, together with the realization of first-layer Pansu differentials by invariant means, yields the multiplicative constant gap stated above.
Read the full abstract ▾
Sound
LiveSynth: A Streaming Neural Synthesizer for Instrument Cloning and Text-to-Instrument
Kyungsu Kim, Yejin Kim and Kyogu Lee
Recently, generative neural synthesizers have enabled zero-shot instrument cloning from reference recordings and timbre generation from text
prompts. However, existing systems generate complete audio segments offline, precluding their use as playable instruments. We present LiveSynth, a streaming neural synthesizer with flexible timbre control and interactive generation capabilities, including zero-shot instrument cloning and text-to-instrument synthesis. A feedback-free causal
Transformer conditioned on MIDI events and a CLAP timbre
embedding outputs continuous audio latents at 100 Hz, which a causal variational autoencoder decodes into audio. Frame-wise timbre conditioning enables continuous timbre morphing, while MIDI-conditioning dropout enables autonomous continuation from a performed prefix. On the NSynth and Slakh
datasets, LiveSynth achieved timbre similarity and spectral reconstruction quality comparable to offline baselines. On an M3 MacBook Air, LiveSynth required 8.8 ms of computation per 10 ms audio frame (an RTF of 0.88) with 10 ms algorithmic latency, enabling real-time interaction with a performer.
Read the full abstract ▾
Computer Vision
UniMergeTrack: Hierarchical Model Merging for General RGB-X Visual Object Tracking
Shiao Wang, Xiao Wang, Yuhe Ding et al. (+5 more)
RGB-X tracking leverages complementary modalities, such as thermal infrared, event, or depth, to improve robustness under challenging conditions. Existing general-purpose RGB-X trackers typically follow two paradigms: independent task-specific training and joint training across modality combinations. However, the former suffers from knowledge isolation and model redundancy, while the latter introduces optimization interference across heterogeneous tasks. In this work, we present UniMergeTrack, a new model merging paradigm that preserves task-specific optimization while enabling effective cross-task knowledge integration. UniMergeTrack first trains modality-specific trackers independently and then performs hierarchical model merging. Specifically, transferable backbones are merged across RGB-X tasks via high-pass-filtered merging, and complemented by lightweight task-specific experts that preserve modality-specific cues. Tracking heads are merged within each task using optimal transport-guided channel
alignment to mitigate hidden channel mismatch. A subsequent distribution
alignment fine-tuning stage further improves compatibility between the unified backbone and task-specific heads. Extensive experiments on the LasHeR, RGBT234, VisEvent, VOT-RGBD22, COESOT, and CDTB
benchmarks demonstrate that UniMergeTrack achieves state-of-the-art or competitive performance across RGB-Thermal, RGB-Event, and RGB-Depth tracking tasks, while exhibiting strong generalization to unseen
datasets. The source code will be released upon acceptance. The source code of this paper will be released on https://github.com/Event-AHU/OpenEvTracking.
Read the full abstract ▾
All authors Shiao Wang, Xiao Wang, Yuhe Ding, Jiarui Xu, Zikang Yan, Bo Jiang, Yonghong Tian, Jin Tang