Sound
Where the Body Keeps the Beat: Structured Motion Conditioning and Music Dynamics Supervision for Dance-to-Music Generation
Changchang Sun, Lu Cheng and Yan Yan
Dance-to-music (D2M) generation aims to synthesize musically plausible soundtracks whose temporal structure aligns with a given dance performance. Representative D2M methods do not explicitly distinguish motion cues across body parts and frequency bands, while standard flow matching lacks a dedicated objective for supervising local music-latent dynamics. To address these issues, we propose Dyna2Music, a latent flow-matching framework that combines structured motion conditioning with explicit supervision of music-latent dynamics. An empirical analysis on AIST++ quantifies how spatial partitioning and frequency separation affect raw music-to-kinematic beat
alignment and motion-reference density, informing the conditioning design. Accordingly, Dyna2Music decomposes joint velocities into slow and fast components and hierarchically fuses the resulting part-wise motion energy with pretrained joint features to condition music generation. To complement this representation, we introduce latent dynamics consistency (LDC), an auxiliary objective that matches adjacent-frame change magnitudes between a single-step clean-latent estimate and the paired reference. LDC makes local music-latent variation an explicit training target without adding trainable parameters or inference computation. Dyna2Music supports variable-length music generation, and experiments on AIST++ and TikTok demonstrate improved rhythmic
alignment and audio quality over representative D2M baselines.
Read the full abstract ▾
Cosmology and Nongalactic Astrophysics
Emulation of the Halo Mass Function with Neural Networks
Olive Ross and Nicholas Battaglia
The abundance of
galaxy clusters as a function of mass — the
halo mass function (HMF) — depends strongly on cosmological parameters. Large-scale simulations are computationally expensive, motivating emulators that interpolate simulation outputs. We present a neural network emulator that maps cosmological and astrophysical parameters,
halo mass, and
redshift directly to a cumulative count of
halos above the input mass, avoiding the information loss inherent to binning and to the functional or PCA-based data reduction used in prior emulators. Each emulator prediction is accompanied by a self-estimated error, expressed in units of Poisson shot noise, that the network learns to predict alongside the
halo count itself — giving every output a built-in confidence metric. We validate the emulator with leave-one-out tests on three simulation suites — the N-body MassiveNus suite and the hydrodynamical CAMELS IllustrisTNG and SIMBA suites — spanning a range of box sizes,
redshifts, sub-grid physics, and softwares. Of all the held-out predictions ($1.2\times 10^6$ in total), 88% have model bias at or below shot noise and 99.9% are within $3\times$ shot noise. The only significant high-error tail is in MassiveNus, where the predicted model bias flags 76% of predictions exceeding $3\times$ shot noise; the remaining number of high error predictions is comparable to the number expected from shot noise alone. This emulator is publicly available and can be trained on other simulation suites making it a flexible HMF tool that can be used for cosmological and astrophysical inference.
Read the full abstract ▾
Computation and Language
Reason in Style: Discovering and Controlling Style in Language Models
Ioana Marinescu, Eric Karl Oermann and Kyunghyun Cho
Language models learn content and style jointly, making stylistic variation in their outputs difficult to identify and control. We study whether recurring styles in model responses can be discovered without supervision and explicitly controlled. We design an
algorithm that learns to separate representations of content and style from
language models' outputs and validate its effectiveness on math questions in a controlled setting. By applying this method to over 100K verified traces from nine distinct teacher models, we discover six recurring yet imbalanced styles. We then fine-tune smaller student models to follow these styles when explicitly conditioned on them, using importance weighting to balance the contribution of the styles represented in the corpus. This approach improves Pass@$k$ over standard fine-tuning on the same data across six math
reasoning benchmarks, demonstrating that we can diversify the style of answers effectively. We confirm that this also results in strong correspondence between requested and realized styles. We find that style affects correctness: the probability of solving a problem depends on the style we condition on, and different problems benefit from different styles. In summary, our results show that stylistic variation in model-generated data can be discovered in an unsupervised way, and made explicit, providing a source of both control and improved
reasoning performance.
Read the full abstract ▾
High Energy Physics (Phenomenology)
Reinaldo Francener and Victor P. Goncalves
A TeV-scale
muon collider enables the production of a highly collimated high-energy
neutrino flux by
muon decay in the straight sections of the
collider ring. Recent studies have demonstrated that these
neutrinos can be used for high-precision measurements in
neutrino physics with a dedicated far-forward
neutrino detector. In this study, we investigate the potentiality of the
muon collider to discover the rare
neutrino trident process, which is characterized by the fusion between a weak
boson from the
neutrino and a
photon from the Coulomb field of the
nucleus target. Our results indicate large
neutrino trident rates per year [$\mathcal{O}(10^3)$] in a baseline 10 kg
neutrino detector for a configuration of different flavour pairs of charged
leptons in the final state, which will allow high-precision tests of the Standard Model and searches for New Physics. As a first example, we investigate the impact of the a new gauge
boson $Z'$, as predicted by the $L_\mu - L_\tau$ model, and derive the projected sensitivity to this new gauge
boson. Our findings show that a
neutrino detector at a
muon collider will significantly extend the current excluded regions of the parameter space.
Read the full abstract ▾
Machine Learning
JEPA-TTT: Persistent Test-Time Training of Latent World Models for Planning under Dynamics Shifts
Zheyuan Zhang, Suyu Ye, Nakul Agarwal et al. (+5 more)
World models enable
agents to plan by predicting future states of the environment, but their predictions can become unreliable when test-time dynamics differ from those seen during training. We present JEPA-TTT, which adapts the latent dynamics predictor of a pretrained action-conditioned Joint-
Embedding Predictive Architecture world model throughout test time. Self-supervised updates accumulate across episodes, while the visual encoder and reward head remain fixed, preserving the pretrained representation and task objective. Planning requires neither a goal image nor online environment reward. JEPA-TTT uses dense replay, which forms prediction windows at every temporal offset, retains them in a growing buffer, and samples minibatches from that buffer for predictor updates. Across eight dynamics shifts in four continuous-control environments, JEPA-TTT improves planning on every shift. After 500 test-time episodes, it reduces autoregressive latent prediction error by 83% on average and improves planning performance by 153% over the frozen JEPA world model. These results show that persistent self-supervised test-time training can adapt a pretrained latent world model under changed dynamics.
Read the full abstract ▾
All authors Zheyuan Zhang, Suyu Ye, Nakul Agarwal, Hossein Nourkhiz Mahjoub, Ehsan Moradi Pari, Daniel Khashabi, Tianmin Shu, Vaishnav Tadiparthi
Audio and Speech Processing
Frequency-Weighted Soft-Constrained Spatially Selective Active Noise Control for Open-Fitting Hearables
Tong Xiao, Reinhild Roden, Matthias Blau et al. (+1 more)
Recently, soft-constrained spatially selective active noise control (SSANC) for open-fitting hearables has been proposed to enable a flexible trade-off between noise reduction and desired-
speech preservation using a scalar trade-off parameter across all frequencies. This paper generalizes soft-constrained SSANC by introducing frequency-dependent weighting of the desired-
speech preservation error while maintaining low-delay time-domain active control. Frequency-dependent weightings based on the
speech intelligibility index (SII), the modified intermediate reference system (MIRS), the long-term average
speech spectrum (LTASS), and a signal-dependent oracle are compared with scalar weighting using measured acoustic impulse responses in a multi-speaker scenario. The results show that LTASS and oracle weightings provide the largest
speech-intelligibility improvements and comparable
speech-quality improvements at substantially lower distortion levels than those obtained with SII and MIRS weightings. While the oracle weighting provides the best overall perceptual trade-off, LTASS weighting provides a comparable perceptual trade-off without requiring the clean desired-
speech signal.
Read the full abstract ▾
All authors Tong Xiao, Reinhild Roden, Matthias Blau, Simon Doclo
Game Theory
Outer Diversity of Condorcet Domains
Piotr Faliszewski, Jan Jabrocki, Mateusz Słuszniak et al. (+3 more)
A Condorcet domain is a set of rankings over a given candidate set, such that every election that consists only of (an odd number of) votes from the domain has a transitive majority relation. We study outer diversity of Condorcet domains, i.e., a measure that quantifies expected swap distance from a random vote to a closest one in the domain. We numerically analyze outer diversity for maximal Condorcet domains with few candidates, and then we establish its asymptotic behavior for several special domains, mostly obtaining theoretical results.
Read the full abstract ▾
All authors Piotr Faliszewski, Jan Jabrocki, Mateusz Słuszniak, Krzysztof Sornat, Stanisław Szufa, Tomasz Wąs
Neural and Evolutionary Computing
Neuromorphic Pseudo-Random Number Generators with a Low Power Hardware Implementation
Jafar Shamsi, Navid Akbari, Sonia Sennik et al. (+2 more)
Pseudo-random number generation often requires trade-offs among quality, power consumption, and bandwidth to produce unpredictable sequences of numbers. The
brain, on the other hand, efficiently generates unpredictable output complex network dynamics occurring in a high-dimensional state. This state, which is hypothesized to be chaotic, relies on the balance between excitation and inhibition. Here, we investigated if computational models of these chaotic balanced states can be harnessed for Neuromorphic Pseudo-Random Number Generators (NPRNGs) in low power hardware. We successfully constructed a balanced spiking neural network model consisting of leaky-integrate-and-fire
neurons that could be readily implemented in low power FPGAs and used as a NPRNG. The prototyped NPRNG consumed 3.24 mW during operation and produced pseudo-random numbers at 120kbps. In both hardware and software instantiations, NPRNGs produce high-quality random numbers as validated by standard metrics for testing RNG quality.
Read the full abstract ▾
All authors Jafar Shamsi, Navid Akbari, Sonia Sennik, Aaron Gruber, Wilten Nicola
Robotics
Toward Humanoid Robots in Construction: A Teleoperation Feasibility Study
Parastoo Ali Pour, David R. Martin, Chang Min Hur et al. (+6 more)
We present a teleoperation system that enables a single operator to perform construction tasks on a Unitree G1
humanoid, combining extended reality (XR) based upper body control with pedal-based locomotion to enable simultaneous manipulation and locomotion. Motivated by persistent labor shortages, hazardous working conditions, and challenges in
humanoid autonomy, we investigate teleoperation as a practical near-term approach for reducing physical strain on workers while generating high quality demonstration data. We evaluate the system on two representative construction tasks drawn from O*NET occupational database, and report task success and completion time relative to a manual baseline. The system achieved 100% success on tool transport and 80% success on surface painting, with teleoperation requiring substantially more time compared to manual execution.
Read the full abstract ▾
All authors Parastoo Ali Pour, David R. Martin, Chang Min Hur, Bo Zhang, Tommy Zhou, Brandon Thomas Lichter, Shane Stanfield, Pramod Khargonekar, Mohammad Abdullah Al Faruque
Computation and Language
Sequential Functional Structured Tucker Compression for Large Language Model Attentions
Jiangfeng Chen, Xinyu Wang, Tianshuo Yan et al. (+4 more)
Post-training compression of LLM
attention is often formulated as independent
matrix approximation, ignoring both the shared structure among
attention projections and the representation shift introduced by earlier compression. We propose FTC, a sequential structured compression framework that adapts the approximation to the current compressed model while jointly exploiting the native Q/K/V head structure under a fixed storage budget. The output projection is handled separately to account for the changed post-
attention representation. FTC requires neither fine-tuning nor gradient-based recovery. Across seven decoder-only LLMs from 6B to 32B parameters, FTC achieves the lowest WikiText-2 perplexity among the compared methods at every tested keep ratio on five modern GQA models, with the largest gains under aggressive compression. The improvements transfer to downstream tasks and remain substantial at the 32B scale.
Read the full abstract ▾
All authors Jiangfeng Chen, Xinyu Wang, Tianshuo Yan, Hanwei Wu, Xiao-Wen Chang, Yang Zhang, Lei Ding
Differential Geometry
Strict Log-Concavity of First Dirichlet Eigenfunctions under Symmetry in $\mathbb{CP}^n$
Jia-Lin Hsu
We prove that the positive first Dirichlet eigenfunction of every smooth domain with convex boundary in $\mathbb{CP}^n$ invariant under the standard $U(1)\times U(n-1)$-action is strictly log-concave for $n=2,3,4$.
Artificial Intelligence
Robust Nash Alignment under Preference Uncertainty
Shihab Ahmed, Debamita Ghosh, David Tang et al. (+3 more)
Preference-based
alignment methods typically optimize against a single preference model, and can therefore be brittle when pairwise preferences are uncertain: noisy, heterogeneous, or shift after deployment. To address these issues, we propose Robust Nash
Alignment, a game-theoretic framework for
alignment to uncertain pairwise preferences. Our formulation has a major learner seeking a policy with a large worst-case win rate against both an
adversarial competitor and any preference kernel lying in an ambiguity set around a nominal preference. When the ambiguity set captures the uncertainty in preferences, the resulting robust objective of the game directly yields a certified lower bound on worst-case performance. However, we note this problem is computationally challenging to optimize, and to address this, we introduce a four-player primal-dual proxy game involving the leader policy, follower policy,
adversarial kernel, and dual variable, and develop a single-loop optimistic mirror descent-ascent
algorithm for it. We show that the proxy always lower-bounds the truncated hard-constrained objective, quantify the proxy-to-hard gap, and characterize an exactness condition under which the proxy recovers the robust objective. We then prove an $\mathcal{O}(1/\sqrt{T})$ average-iteration convergence for the proxy-game duality gap, which implies a near-optimal robust policy for the original robust objective. Experiments on controlled tabular games and LLM
alignment with uncertain preference illustrate the method and show improved robustness to preference perturbations over nominal baselines.
Read the full abstract ▾
All authors Shihab Ahmed, Debamita Ghosh, David Tang, Yudan Wang, Alvaro Velasquez, Yue Wang
Distributed Computing
MANTA: Machine Learning Augmented Tiering Advisor
Johannes Freischuetz, Kiet Pham, Sujay Yadalam et al. (+3 more)
Memory tiering has been used to expand
memory capacity, particularly in datacenters, by combining fast DRAM with slower tiers, including CXL-attached
memory. Its effectiveness depends on keeping useful pages in the fast tier, but existing heuristic policies can lag behind changing hot sets in phased or bursty workloads. To explore these limitations, we introduce ChOMP, a scalable offline optimizer that minimizes placement and bandwidth-sensitive migration costs. We then develop a trace-driven simulator that uses this reference to identify performance opportunities for online policies. Motivated by these results, MANTA predicts future page usefulness from runtime access features and integrates a lightweight learned model into ARMS. Across eight workloads on emulated CXL, MANTA achieves
geometric-mean speedups over ARMS of 1.12$\times$ and 1.08$\times$ at 4~GB of fast
memory on Linux 6.2 and 6.18, respectively; across six Optane workloads, it achieves 1.69$\times$. On individual workloads, MANTA is up to 1.25$\times$ faster than ARMS with emulated CXL on Linux 6.2, 1.21$\times$ on Linux 6.18, and 5.6$\times$ with Optane.
Read the full abstract ▾
All authors Johannes Freischuetz, Kiet Pham, Sujay Yadalam, Konstantinos Kanellis, Michael Swift, Shivaram Venkataraman
Machine Learning
WOMBAT: Whitebox Oracle for Molecular Benchmarking and Attribution Testing
Dominik Matuszek, Bartosz Zieliński, Tomasz Danel et al. (+1 more)
When a graph neural network (GNN) explainer produces an unexpected attribution on a molecule, the attribution alone cannot reveal whether the explainer has failed or the model has learned a shortcut. We introduce WOMBAT, a
benchmark of 14 whitebox GNNs, each with message-passing weights set by hand to detect a specific SMARTS motif. Each model's decision rule is known by construction, providing attribution ground truth against which explainer errors can be identified and studied. We validate the models on millions of PubChem molecules and evaluate post-hoc explainers including GNNExplainer, PGExplainer, and Integrated Gradients. Guided by our qualitative analysis, we construct a model that causes Integrated Gradients to spread attribution across the graph, even though the model reliably detects the intended motif. We release the
dataset, models, and evaluation code to help researchers in the development of newer XAI tools for GNNs.
Read the full abstract ▾
All authors Dominik Matuszek, Bartosz Zieliński, Tomasz Danel, Dawid Rymarczyk
Quantum Gases
Krylov complexity of a tilted extended Bose-Hubbard chain with Rydberg-dressed interactions
Yifan Chen, Tianyi Yan, Lu Qin et al. (+1 more)
We investigate Krylov state complexity in a tilted extended Bose-Hubbard chain in which both the on-site interaction $U$ and nearest-neighbor interaction $V$ are present. The tilted extended Bose-Hubbard model can be realized with Rydberg-dressed interactions in optical
lattices. Using exact diagonalization and Lanczos recursion from three physically motivated Fock states, we compute the time-dependent complexity, its long-time saturation value, and the fluctuations of the Lanczos coefficients for an open chain at unit filling. Crucially, we identify a pronounced quasi-chaotic diagonal regime near $U\simeq V$ in which the Krylov complexity saturation exhibits a sharp ridge while the spectral level-spacing statistics remain partially chaotic. This dissociation arises from the competition between the on-site and nearest-neighbor interaction, which generates a dense
manifold of near-resonant many-body configurations that is efficiently coupled by the state-dependent dynamics but retains residual spectral structure. Our results show that Krylov state complexity provides complementary, state-sensitive information about many-body chaos that is not captured by the spectral diagnostics alone, establishing Krylov complexity as a useful diagnostic for studying thermalization and information scrambling in experimentally accessible
quantum simulators.
Read the full abstract ▾
All authors Yifan Chen, Tianyi Yan, Lu Qin, Weibin Li
Statistical Methodology
Discussion on "Principled type I error rate inflation in two-arm clinical trial designs with external control information borrowing" by Silvia Calderazzo, Manuel Wiesenfarth, Vivienn Weru, and Annette Kopp-Schneider
Tim Friede and Christian Röver
Consideration of frequentist type-I error or power based on certain fixed conditions may not fully embrace the aims of a
Bayesian analysis; this is particularly true in settings where informative priors are involved. Calderazzo et al. consider adaptations of a testing procedure involving
Bayesian borrowing from external control data. Particular caution may be necessary here in order not to condition on parameters and data simultaneously. We consider their example case and argue that (a) the conditioning scheme may have limited practical implications, and (b) if the main concern are assumption violations, these might be better addressed by adapting the likelihood function.
Read the full abstract ▾
Artificial Intelligence
ReLiveGym: Evaluating Long-Lived Agents over Weeks of Replayed Reality
Xisen Jin, Jingheng Li, Zhenglun Chen et al. (+2 more)
As large
language model (LLM)
agents become widely adopted, they are increasingly deployed for tasks that require persistent monitoring or recurring actions (e.g.,
market analysis). These
agents are expected to operate unattended for days or weeks, act at the right timing, and adapt to the dynamic environment over time. These challenges are not fully captured in the existing long-horizon
agent work, as they often consider a static environment that is not temporally changing. We introduce ReLiveGym, a diagnostic evaluation environment of long-lived tasks in which
agents act sparsely over simulated weeks of chronologically replayed real-world news,
market, and social-media streams. The tasks span diverse levels of time sensitivity,
reasoning intensity, and recurrence. Across eight base
language models, we investigate how model choice and harness design affect
agent performance on such long-lived tasks. Our results show that how
agents determine when to act arises as an important harness-design axis for long-lived tasks; and that the optimal design varies across tasks and sometimes model choices as well. We also evaluate how continuous learning from hindsight feedback affects performance and addresses failure modes observed in these long-lived tasks. These findings indicate model choice, action timing mechanism, and use of feedback as important considerations in the design of long-lived
agents. Code: https://github.com/SaharaLabsAI/ReLiveGym
Read the full abstract ▾
All authors Xisen Jin, Jingheng Li, Zhenglun Chen, Junyi Du, Xiang Ren
Algebraic Geometry
Automorphisms of hypersurfaces in polarized varieties
Ana Quedo
Let $F$ be a projective variety, let $H$ be a normally generated line bundle on $F$, and let $X\in |H^{\otimes d}|$ be a general projective variety. We establish a uniform criterion, formulated in terms of the graded Betti numbers of the section ring $R(F,H)$, that provides sufficient conditions for every
automorphism of $X$ to extend to an
automorphism of $F$. This extends the classical results of Matsumura-Monsky on
automorphisms of hypersurfaces in projective spaces to a broader class of polarized varieties $(F,H)$, including rational homogeneous varieties of Picard number one, smooth Fano threefolds with their anticanonical polarization, and examples of hyperkähler and Calabi-Yau varieties.
Read the full abstract ▾
Machine Learning
Beyond Unimodal Bases: Pullback Geometry for Multimodal Data
Honglei Brinkmann, Lucas Ng, Georgios Batzolis et al. (+3 more)
Data-driven Riemannian geometry provides nonlinear interpolation and
geometric representations of high-dimensional data. For these operations to be statistically meaningful, paths between observations should preferentially traverse high-likelihood regions. Existing scalable pullback constructions typically use a unimodal Gaussian latent distribution, assuming that the data reside close to a single
manifold. For
multimodal data, mapping separated modes or local structures into one Gaussian region can require substantial transport deformation and compromise the resulting geometry. We introduce a pullback geometry for data supported on mixtures of
manifolds. Using a latent Gaussian mixture, we define its Riemannian metric as the
matrix square of the responsibility-weighted expected component precision. The metric is smooth and positive definite and recovers the existing Gaussian construction in the single-component limit. For structured overlapping mixtures, we establish conditions under which the log-density is concave along geodesics, providing a formal connection between the proposed geometry and paths through high-likelihood regions, and derive the corresponding local
curvature relations. We instantiate this geometry in a normalizing flow with adaptive mixture learning, allowing the number of active components to emerge from the data and supporting component-wise reconstruction and local effective-dimension estimation. Experiments on synthetic
geometric data, a controlled multi-view image setting with a known reference trajectory, and MNIST show reduced transport distortion, competitive path support, close reference-trajectory recovery, and improved interpolation realism. These results extend scalable pullback geometry beyond
datasets that reside close to a single
manifold while retaining tractable and interpretable local structure.
Read the full abstract ▾
All authors Honglei Brinkmann, Lucas Ng, Georgios Batzolis, Mark Girolami, Carola-Bibiane Schönlieb, Willem Diepeveen
Machine Learning
Initialization Improves LLM-Driven Discovery
Mansi Sakarvadia, Marco Ciccone and Colin Raffel
Large
Language Models (LLMs) have been used for novel discovery of
algorithms,
theorems, drugs, and other tasks through the use of harnesses that
prompt an LLM to iteratively optimize an objective. In this work, we study the relationship between the population of previous iterates and eventual discovery success. We generalize past work on harness design to develop a suite of 12 harnesses called 'Modular' and characterize their performance across 5 diverse discovery tasks, finding that discovery success is brittle and sensitive to harness design. We uncover mode collapse, characterized by a dramatic drop in the diversity of iterates, as a common failure mode. We find that popular state-of-the-art harnesses and diversity-inducing harness interventions, which aim to prolong this collapse, yield inconsistent gains. Our results instead uncover that the performance of early discoveries is predictive of eventual success. We therefore propose a universally applicable intervention that performs an initial stage of parallel exploration in order to initialize subsequent iterative optimization. Our method provides consistent gains across many harnesses and target applications, confirming the importance of initialization in LLM-driven discovery.
Read the full abstract ▾
Sound
AnchorPrompt: Self-Distilled Soft Prompts for Robust Audio-Language Models
Pooneh Mousavi, Amir Ivry, Mirco Ravanelli et al. (+1 more)
Large audio-
language models (LALMs) are sensitive to input perturbations, such as noise, waveform corruption, and
adversarial injections. We propose AnchorPrompt, an efficient adaptation method that keeps the model frozen and learns a single block of
prompt vectors inserted at the decoder input, between the audio and question
embeddings. We train these vectors through self-
distillation over diverse audio and text perturbations. To improve answer consistency and mitigate
hallucination, we use the model's prediction on the clean recording as the target for answerable inputs, and assign a refusal target when the audio lacks sufficient evidence to answer. Furthermore, AnchorPrompt is perturbation-agnostic at inference, requiring no prior detection of perturbations and enabling zero-shot transfer to unseen distortions. We evaluate three LALMs across three
benchmarks and show that AnchorPrompt improves answer consistency in most tested conditions. Clean accuracy improves in six of nine model-
benchmark pairs, with minimal impact on the remainder of 1.2% at most. Crucially, AnchorPrompt reduces
hallucinations under severe audio corruption while keeping false refusals on clean audio rare. Finally, these consistency gains transfer to unseen perturbations, such as choice permutations and reverberation.
Read the full abstract ▾
All authors Pooneh Mousavi, Amir Ivry, Mirco Ravanelli, Cem Subakan
Artificial Intelligence
Meta-Multi-Agent Reinforcement Learning for Fast Adaptation of Interactive Policies with Applications to Autonomous Driving
Huiwen Yan, Kyriakos G. Vamvoudakis and Mushuang Liu
This paper develops a meta-multi-
agent reinforcement learning (meta-MARL) framework to enable fast adaptation of interactive policies in a multi-
agent system (MAS). Meta-reinforcement learning (meta-RL) enables
agents to rapidly adapt to new tasks/environments using a bi-level optimization mechanism. However, existing meta-RL generally focuses on single-
agent systems. Extending these frameworks and
algorithms to multi-
agent systems poses additional challenges, as tasks are characterized by not only the environment but also
agents' strategic interactions. To address these challenges, we model multi-
agent reinforcement learning (MARL) problems as Markov games (MGs) and develop a meta-MARL framework for rapid interactive policy adaptation across a distribution of MGs. A new concept, called meta-NE, is defined to describe the desired solution concept in a meta-MARL problem. Sufficient conditions for the equivalence between a meta-NE and a stationary point of the gradient-play-based meta-MARL
algorithm are established. Our evaluation on autonomous-driving tasks demonstrates that the proposed meta-MARL method achieves faster adaptation than pretrained MARL baselines, validating the effectiveness of our framework.
Read the full abstract ▾
Machine Learning
SkillSpec: Consensus-Gated Agent Skill Evolution via Representation Specialization
Huancheng Chen, Xiaodi Sun, Zhaoqiong Huang et al. (+3 more)
Natural-
language skills are textual procedural
memories through which large
language model (LLM)
agents retain reusable task knowledge without updating model weights. Existing methods typically treat skills as either static artifacts or monolithic documents optimized using aggregate validation scores as feedback. However, representing a skill as a monolithic document restricts optimization to its textual content, without explicitly modeling the structure through which procedural knowledge is retrieved and executed. We identify a key distinction between learning what knowledge to retain and determining how to organize it: textual updates should first be validated through execution evidence, after which the retained knowledge should be structured according to its procedural dependencies and
retrieval requirements. To this end, we introduce SkillSpec, a two-phase framework comprising consensus-gated
evolution and representation specialization. In the consensus-gated phase, complementary editing intents generate complete candidate skills. An update is committed only when paired evaluations reach consensus, requiring sufficient overall improvement and non-negative aggregate paired gain in every repeated evaluation. In the specialization phase, signals of process and redundancy sensitivity derived from the full optimization trajectory, including accepted and rejected candidates, guide the selection of a flat, graph, or hybrid representation.Across six
benchmarks and three target
language models, SkillSpec improves average success rate over SkillOpt by 6.89%, averaged across the three models. These results demonstrate that reliable skill
evolution and representation specialization address complementary objectives: deciding what knowledge to retain and how to structure it for inference.
Read the full abstract ▾
All authors Huancheng Chen, Xiaodi Sun, Zhaoqiong Huang, Shenyang Huang, Shreya Singhal, Jingwen Lu
Functional Analysis
Riesz transform characterization of Hardy spaces associated with Dunkl–Schrödinger operators
Agnieszka Hejna-Łyżwa
Let $L=-\Delta_k+V$ be a Schrödinger operator associated with the Dunkl
Laplacian and a nonnegative potential. We characterize the Hardy space $H_L^1$, defined by the maximal operator of the
semigroup generated by $-L$, in terms of the Riesz transforms associated with $L$. Under a local integral condition on the potential and a decay condition on the
semigroup, formulated relative to an admissible covering by cubes, we prove that \[ \|f\|_{H_L^1} \asymp \|f\|_{L^1(dw)} +\sum_{j=1}^N \|\widetilde{R}_j f\|_{L^1(dw)}. \] The proof combines an atomic characterization of $H_L^1$ with local Hardy space theory and weighted integral estimates for Dunkl derivatives of the heat kernels. The result applies to arbitrary root systems and nonnegative multiplicity functions.
Read the full abstract ▾
Distributed Computing
Ditto: Generalized Reconfigurable Linearizable Reads
Aleksey Panas
Linearizability gives developers the illusion that operations on a distributed datastore execute sequentially on a single machine. Providing it is expensive, so numerous specialized reads algo- rithms have been proposed to make reads significantly faster without sacrificing linearizability, which matters because most workloads are read-dominant by orders of magnitude. Yet, none of these
algorithms is the best choice for every deployment. We show why with a framework that classifies these
algorithms into a generalized space together with a mathematical analysis of their latencies as a function of this generalization. The analysis exposes two fundamental tradeoffs, one between read and write latency and one between read and write tolerance to network variance, which together leave the best
algorithm dependent on network and workload conditions that vary over time. It also exposes a gap: there is a way to minimize the delay a read incurs when it overlaps a concurrent write, at no cost to writes, yet exist- ing
algorithms can apply it only for a subset of the tradeoff space, forcing any deployment that needs the full space to settle for a strictly worse solution. We close this gap with a new
algorithm, Pairwise Quorums, and build a second, Ditto, that reconfigures across the space at runtime as conditions change. Together they subsume every existing
algorithm under our model: against each, the combination is either strictly better, or reconfigures to match it where a genuine tradeoff applies.
Read the full abstract ▾
Human-Computer Interaction
From Images to Tasks: Characterizing Multimodal LLM Interactions in the Wild
Jinyi Ye, Scott Counts, Gaurav Verma et al. (+2 more)
Multimodal large
language models (LLMs) increasingly integrate
vision and text, yet how people use them in natural settings remains underexplored. We seek to answer the question: when users upload images, what tasks are they trying to accomplish? Analyzing over 40,000 de-identified image-upload conversations from Microsoft Copilot, we characterize real-world
multimodal use through a hierarchical framework of ten capabilities, spanning perception, cognition, and generation. First, we characterize the distribution and composition of these capabilities, finding that the majority of image-upload tasks involve multiple capabilities. Second, we find that
multimodal use spans a broader and more diverse task space than text-only interactions, with asymmetric coverage and task classes that rely on cross-modal grounding. Third, mapping observed capability demand onto 253 existing
benchmarks reveals uneven
alignment between
benchmark coverage and real-world use:
benchmarks concentrate on perception and
reasoning toward fixed answers, while common workflows involving text, code, and data generation remain comparatively undertested. We validate our taxonomy and findings on an independent ChatGPT
dataset. Our results provide a large-scale empirical characterization of what users seek to accomplish with
multimodal LLMs and highlight opportunities for
benchmark design grounded in observed user demand.
Read the full abstract ▾
All authors Jinyi Ye, Scott Counts, Gaurav Verma, Kate Lytvynets, Weiwei Yang
Artificial Intelligence
R-GroundBench: A Diagnostic Benchmark for R-Group Groundingin Markush Molecular Editing
Xin Wang, Zichuan Ying, Xinna Lin et al. (+8 more)
Recent advances in AI for scientific discovery enable molecular understandingand design, yet
reasoning over incomplete chemical representations remainsunclear.Markush structures, which encode molecular families through variable R-groupplaceholders (\textit{R\textsubscript{1}}, \textit{R\textsubscript{2}}, X, etc.), are ubiquitous in pharmaceutical patents and requiregrounding across molecular, textual, and chemical information.However, existing molecule-
language benchmarks focus on fully specifiedmolecules, leaving R-group grounding largely unevaluated.We introduce R-GroundBench:, a diagnostic
benchmark built from real patent Markushstructures, featuring a Multiple-Choice (VQA) track with controlled difficultyand modality splits, and an open-ended Generation track.Our results reveal a substantial gap between recognition andmolecular grounding.While models achieve over 90% accuracy on Easy VQA, performance drops to56–66% on Hard VQA when shortcuts are controlled.Chemical-domain VLMs also remain unreliable, achieving only 25.7–46.2% on HardVQA despite domain-specific pretraining.Moreover, Generation Exact Match remains below 20% for most models and below8% when visual input is required.These findings reveal that current AI systems lack reliable grounding andexecution for Markush editing, highlighting challenges for AI-drivenscientific discovery.
Read the full abstract ▾
All authors Xin Wang, Zichuan Ying, Xinna Lin, Junqi Zhang, Hanyi Xiong, Tianyu Gao, Hairong Zhang, Qixiang Hua, Botian Shi, Zhenhailong Wang, Kaicheng Yu
Nuclear Theory
Coexistence of strange quark stars and neutron stars: metastability and nucleation in proto-neutron stars
Mirco Guerrini, Giuseppe Pagliara, Luca Passarella et al. (+1 more)
If strange
quark matter (SQM) is absolutely stable, hadronic
neutron stars (NSs) and strange
quark stars (QSs) may coexist in the so-called two-families scenario. A key issue is explaining how hadronic NSs can survive as long-lived metastable objects, rather than promptly converting into stable QSs. Since all NSs are born as hot proto-
neutron stars (PNSs), a necessary condition for their existence is that the conditions for conversion, triggered by the nucleation of the first critical SQM droplet, are not reached in every PNS. We investigate the thermal nucleation of critical SQM droplets under representative PNS conditions, accounting for
flavor-composition fluctuations and finite-size effects on color superconductivity by suppressing pairing in droplets smaller than the diquark coherence length. We define the nucleation conditions as the thermodynamic state at which a critical SQM droplet is expected to nucleate statistically within the characteristic dynamical timescale of the system. Under these conditions, the reduced nucleation barrier is nearly universal, with $W_*/T\simeq167$ to within a few units. Exploring the SQM parameter space, we show that requiring a canonical $1.4~\rm{M}_{\odot}$ NS to survive the PNS stage sets a lower bound on the
hadron-
quark surface tension $\sigma\gtrsim 67$ MeVfm$^{-2}$.
Read the full abstract ▾
All authors Mirco Guerrini, Giuseppe Pagliara, Luca Passarella, Alessandro Drago
Statistical Computation
Multifidelity Formulations for Triangular Transport
Owen Davis, Daniel Sharp, Youssef Marzouk et al. (+1 more)
We develop multifidelity methods for constructing triangular transport maps from samples, when high-fidelity data are scarce but lower-fidelity data are more abundant. Using this set of multifidelity data, we approximate a triangular transport map that bijectively maps between a tractable reference density and the high-fidelity target distribution. We introduce two strategies to leverage low-fidelity data: a hierarchical approach that composes maps between adjacent fidelity levels, and a non-hierarchical method that incorporates low-fidelity information through monotonicity-preserving corrections to the map parameterization. Numerical experiments compare these strategies with single-fidelity transport and demonstrate how the proposed multifidelity approaches can improve map estimation from limited high-fidelity data. To illustrate the broader utility of the learned maps, we also deploy them in a downstream amortized simulation-based inference task. This example shows that multifidelity improvements in map estimation can translate to improved conditional sampling and uncertainty quantification when high-fidelity data are scarce.
Read the full abstract ▾
All authors Owen Davis, Daniel Sharp, Youssef Marzouk, Gianluca Geraci
Game Theory
Bandwidth Fee Mechanisms for Certified Transaction Dissemination
Eleftheria Fassman, Mahimna Kelkar, Benjamin Marsh et al. (+1 more)
We study bandwidth fee mechanisms (BFMs) for pricing threshold-certified transaction dissemination before consensus. Analogous to transaction fee mechanisms for computation, BFMs price blockchain communication, which is especially relevant when dissemination is separated from consensus and execution. We model BFMs as two-sided procurement auctions between users submitting transactions and validators receiving and forwarding them. Any validator may serve as a transaction's source and forward it to a threshold of attesters. The model accounts for receipt from the user and subsequent transfers between validators, while abstracting away unrestricted multi-hop gossip and network
topology. We establish limits on myopic incentive compatibility: no nontrivial transcript-based BFM prevents profitable collusion between two validators, and no nontrivial BFM simultaneously satisfies incentive compatibility for individual users, individual validators, and coalitions of one validator and one user. On the feasibility side, we give a greedy-route posted-price mechanism that is incentive compatible for every individual user and validator. We also study bandwidth pricing across blocks. Assuming honest validator execution, we propose an EIP-1559-style price-update rule whose fluid approximation has a unique fixed point that is locally stable for sufficiently small step sizes. In large
markets, prices starting near this point remain nearby with high probability over any fixed time horizon.
Read the full abstract ▾
All authors Eleftheria Fassman, Mahimna Kelkar, Benjamin Marsh, Ke Wu