Robotics
Fast Planning for Multi-object Multi-target Throwing
Zhengming Zhu, Yang Liu, Xiao Gao et al. (+1 more)
Robot throwing has emerged as a promising technique for improving efficiency in logistics and warehouse automation, by enlarging the workspace and speeding up the process. To significantly increase the throwing system's throughput, we develop strategies for throwing multiple objects in one swipe. Such multi-object multi-target throwing (MOMT) leverages the large degrees of freedom of anthropomorphic hands. The key is to quickly generate fast and feasible throwing motions, which involves a complex trade-off between short trajectory duration and short planning time. We solve this problem in two stages. Offline, we build a model of the feasible set by combining object's inverted flying dynamics and the
robot's kinematics and dynamics. Online, we generate feasible throws through fast solution matching and filtering of object's valid detach state and
robot's feasible state that can compose sequences of throws in less than 5 ms. We validate the framework on a 7-DoF manipulator equipped with a multi-fingered hand. In simulation, coordinated two-object throwing reduces execution time by up to 46% compared to independent single-object planning, and this improvement is maintained when scaling to three objects. Real-world experiments with two objects confirm a 29% reduction; the remaining gap to the theoretical 50% is attributed to inter-throw transition overhead. When target positions are randomly changed mid-execution, the system re-plans and successfully reaches the new targets within 100 ms latency without stopping the
robot. These results establish the first unified planning framework for MOMT throwing – demonstrating scalability to multiple objects in simulation and real-world feasibility on two-object tasks – advancing the frontier of high-throughput robotic manipulation. A
video summarizing the method and the hardware experiments is available at https://liuyangdh.github.io/momt-
video
Read the full abstract ▾
All authors Zhengming Zhu, Yang Liu, Xiao Gao, Aude Billard
Computer Vision
PVSync: A Unified Lip-Sync Expert for Timing and Articulation
Kevin Stephen, Varun Menon, Timo Mertens et al. (+1 more)
Lip movements can match the timing of
speech without matching the spoken sounds. We introduce PVSync, a unified model for audio-visual offset estimation and phoneme-level articulation scoring. PVSync combines window-level contrastive learning for synchronisation with a phoneme-level articulation objective that aligns audio and
video embeddings of the same viseme class across clips. Visemes group phonemes with similar visible articulation. Viseme labels are derived automatically from forced-aligned transcripts, without manual annotations. On offset-corrected
videos from 13 talking-head
video generation models, PVSync matches human rankings of lip-sync quality more closely than LSE-C, achieving a Spearman correlation of 0.83 versus 0.34. On an automatically constructed
benchmark from held-out
speech, PVSync distinguishes viseme-matched from mismatched audio-visual pairs with an ROC AUC of 0.91. PVSync also outperforms SyncNet and MTD-VocaLiST in temporal offset recovery on held-out in-the-wild clips. Code and
benchmark data will be released upon acceptance.
Read the full abstract ▾
All authors Kevin Stephen, Varun Menon, Timo Mertens, Nikita Drobyshev
Materials Science
Quasicrystalline oxide interfaces by twisting SrTiO3 membranes
Shivasheesh Varshney, Sooho Choo, Huan Liu et al. (+34 more)
Quasicrystals reveal that long-range order can exist without translational periodicity, but deterministic routes for designing strongly bonded quasicrystalline interfaces remain unexplored. Here, we show that such interfaces can be engineered by growing epitaxial SrTiO3 membranes on a sacrificial layer, releasing them, twisting them by prescribed angles, and bonding them onto a chosen substrate. SrTiO3
bilayers twisted by 45{\deg} form an octagonal quasicrystalline interface with eight-fold rotational
symmetry, whereas trilayers assembled with successive 30{\deg} rotations form a dodecagonal interface with twelve-fold
symmetry. Selected-area and nano-beam electron diffraction, multislice electron ptychography, atomic-resolution dark field imaging,
Fourier analysis, and real-space tiling establish long-range quasicrystalline order with electronic reconstruction evident from electron energy-loss spectroscopy after annealing. Atomistic simulations based on machine-learned interatomic potentials show that octagonal order emerges as the
bilayer twist angle approaches 45{\deg}. Polarization-dependent X-ray absorption spectroscopy in 1.5 nm twisted
bilayers reveals a twist-dependent electronic response, and terahertz-field-induced second-harmonic generation circular dichroism identifies a
symmetry-dependent nonlinear optical response at the 45{\deg} quasicrystalline condition. Twisted
oxide membranes therefore provide an extendable route to engineer quasiperiodic order and couple it to interfacial bonding, orbital reconstruction, and nonlinear optical response in complex
oxides.
Read the full abstract ▾
All authors Shivasheesh Varshney, Sooho Choo, Huan Liu, In Hyeok Choi, Nicolas Gauquelin, John Mangeri, Seung Gyo Jeong, Edwin Dollekamp, William Sandholt, Andrea Roberto Insinga, Nikolas Vitaliti, Bridget R. Denzer, Bonnie Y. X. Lin, Andrew T. Danbury, Majid Mohseni, Uditha Mihiranga Jayathilake, Sharup Sheikh, Lidia A. Lapinski, Jonas M. Hansen, Seungjun Lee, Dong Kyu Lee, Jitin Sathish Kumar, Shivam Sharma, Rishi Raj, Jay Shah, Alevtina Smekhova, Florian Kronast, Juan Maria García Lastra, K. Andre Mkhoyan, Tony Low, James M. LeBeau, Alexander X. Gray, Jo Verbeek, Keith A. Nelson, Nini Pryds, Richard D. James, Bharat Jalan
Machine Learning
Consistent Distribution Matching for Data-Free Diffusion Distillation
Yuxiang Fu, Qi Yan, Zike Wu et al. (+4 more)
Flow and
diffusion models suffer from slow inference due to computationally expensive numerical integration.
Distillation provides a promising way for a student model to learn from a teacher's dynamics, enabling one-step or few-step generation. However, existing methods often depend on curated
distillation datasets, costly teacher rollouts, or auxiliary proxy networks, which complicate model training and scaling. In this work, we propose Consistent Distribution Matching, a simulation-free and data-free
distillation method for accelerating
diffusion and flow models while preserving strong generative capacity. Our key insight is to unify sample generation and score estimation with one student network. Thus, our framework uses only two models, a frozen teacher and a trainable student, and optimizes one objective. We prove that minimizing our objective indicates Wasserstein convergence of the student flow-map pushforwards to the teacher marginals. On ImageNet 256$\times$256, our method attains an FID of 2.04 with a single function evaluation (1-NFE) and a 4-NFE FID of 1.37 within 40 epochs of training, surpassing the state-of-the-art
distillation baselines without data. Our code code and model are available at https://consistentdmd.github.io/.
Read the full abstract ▾
All authors Yuxiang Fu, Qi Yan, Zike Wu, Yongxing Zhang, Purang Abolmaesumi, Lele Wang, Renjie Liao
Optimization and Control
Regret-Optimal Linear-Quadratic Control under Distributional Drift
Lukas-Benedikt Fiechtner and Jose Blanchet
We study finite-horizon LQR with independent disturbances whose laws may drift over time. We address uncertainty in this
evolution through ex-ante distributionally robust regret optimization (DRRO), which minimizes worst-case excess expected cost relative to the optimal causal controller that knows the law sequence. To specify the admissible law sequences, we introduce a temporally coupled moment ambiguity set with a Gelbrich kinetic-energy budget and a second budget limiting departures from reference moments and persistent cumulative mismatch. Despite the coupling of moments across time, affine disturbance-feedback synthesis reduces exactly to a semidefinite program. An optimal policy uses past disturbances to predict the correction to nominal LQR determined by current and future disturbance means. The SDP solution also yields worst-case moment paths and corresponding laws. For standard distributionally robust optimization (DRO), the analogous SDP relaxation can have a duality gap. Portfolio liquidation experiments show substantially lower worst-case regret over the drifting ambiguity set than common-law DRRO and nominal control.
Read the full abstract ▾
Artificial Intelligence
CM-DPO: Constraint-Margin Direct Preference Optimization for LLM Planning
Rabimba Karanjai, Qun Gu, Hemanth Hegadehalli Madhavarao et al. (+8 more)
Direct Preference Optimization (DPO) treats all constraint violations equally: a $1 budget overshoot and a $1,000 overshoot induce the same training signal. It is also susceptible to length and style bias when preference pairs come from different model families. We introduce Constraint-Margin DPO (CM-DPO), which replaces DPO's binary preference signal with a continuous margin derived from a deterministic symbolic verifier and scaled by violation severity. Hard and soft constraints are separated through a lexicographic objective, ensuring hard constraints are never traded off against preferences. To supply CM-DPO with bias-reduced training pairs, we generate preference data through procedurally generated constraint profiles (DCCG) and minimal-edit
distillation from a
reasoning teacher (RT-MED), within a framework we call SynPlan-R. On TravelPlanner, NaturalPlan, and out-of-distribution PlanBench, an 8B model fine-tuned with CM-DPO achieves 89.2% pass rate and 93.4% solve rate, matching multi-
agent systems at 13x lower latency while outperforming GPT-4o on unseen Blocksworld by 9.2 points.
Read the full abstract ▾
All authors Rabimba Karanjai, Qun Gu, Hemanth Hegadehalli Madhavarao, Wenhuan Sun, Xiaojiao Yu, Suryabhan Singh Hada, Libin N. George, Uma Kona, Richard Williamson, Linsey Pang, Prakhar Mehrotra
Artificial Intelligence
RLDISCOVER: LLM-driven co-evolution of reinforcement learning algorithms
Haoran Li, Zengle Ge, Xiaomin Yuan et al. (+14 more)
LLM-guided program
evolution has enabled discoveries in mathematics and computational optimization, raising the prospect of reinforcement learning (RL)
algorithms that self-evolve to improve how
agents learn. However, realizing this prospect faces two obstacles. Joint search over coupled algorithmic components is difficult to scale: simultaneous changes can disrupt learning, while isolated changes overlook their dependencies. Evaluating candidate
algorithms also requires costly training, with fitness remaining uncertain across random seeds. We introduce RLDiscover, a framework for the self-
evolution of model-free deep RL
algorithms. Progressive Co-
Evolution advances from targeted component edits to joint
evolution, while Progressive Probabilistic Evaluation balances search breadth and evaluation fidelity through staged training and repeated evaluation. Experiments across SAC, PPO, and DQN on four
benchmark suites show substantial improvements in mean return, with per-family median gains of 32%-84% and a peak return ratio of approximately 363x over a near-zero baseline. These gains include transitions from failed learning to successful task completion, and improvements persist when
evolution starts from stronger open-source implementations. On measured SAC locomotion runs, evaluation uses approximately one-fifteenth the estimated compute required to fully evaluate the same candidate pool. Remarkably, independent searches repeatedly discover interpretable combinations of adaptive robust losses, progress-dependent value targets, and running statistics, with selected programs transferring to unseen tasks. These findings point toward a broader role for self-
evolution in AI: discovering interpretable
algorithms that improve how
agents learn.
Read the full abstract ▾
All authors Haoran Li, Zengle Ge, Xiaomin Yuan, Yui Lo, Songlin Zhou, Jiahua Ying, Haoxin Li, Qianhui Liu, Yuanhang Liu, Jiaqun Liu, Guokai Chen, Mingju Chen, Ruinan Wang, Annan Li, Jianmin Wu, Dawei Yin, Dou Shen
Computer Vision
A Deterministic Evidence Layer for Vision-Language Autism Screening from Naturalistic Home Video
Wenqi Li, Mindi Ruan, Chuanbo Hu et al. (+2 more)
Autism spectrum disorder (ASD) is diagnosed through specialist observation of a child's social behavior, and access to that expertise is the bottleneck for early identification.
Vision-
language models (VLMs) describe a child's behavior from
video well; the verdict drawn from the description is unstable: at temperature~0, across eight pipeline configurations on one backbone, 16–37% of clips change their predicted label between repeated runs, and the cause lies in the serving stack. We keep the VLM frozen and move the decision out of the model. Under our grounded perception constraints the VLM writes an event table of timestamped, glossary-labeled events that names the eliciting press and logs counter-evidence; a text-only stage age-calibrates the confidence of each row; a deterministic weight-of-evidence scorer sums it into an evidence total and stratifies it into a risk category, so every decision decomposes into named per-feature contributions and can be re-scored from the saved table. On 43 caregiver-recorded, protocol-free home free-play clips of preschool children, the pipeline reaches AUC $0.851 \pm 0.012$, 86.0% accuracy, and F$_1$ 71.8 over three runs. It labels 74.4% of clips correctly in every run (60.5% for the zero-shot baseline) and flags no typically developing clip in every run (9 of 31 at zero-shot). An ablation on the same backbone attributes the gain to the grounded perception constraints read through the deterministic scorer.
Read the full abstract ▾
All authors Wenqi Li, Mindi Ruan, Chuanbo Hu, Shuo Wang, Xin Li
General Relativity and Quantum Cosmology
Logarithmic corrections to massive particle trajectories for the classical soft log theorems at leading order in $G$
T. H. Nadai
We study the logarithmic corrections to the asymptotic trajectories of massive particles. These corrections enter the calculation of the soft-log sector of the radiative field at null infinity in four-dimensional gravitational scattering. We first isolate the part of the metric which enters the particle equations of motion. We show that the nonlinear source can only affect this sector through homogeneous modes on the hyperboloid, and that matching to the proposed soft tower implies that these modes vanish, except for the first mode. We show that this mode cannot affect the trajectory coefficients at leading order in $G$. We then invert the matter energy-momentum tensor on the particle worldlines to obtain the trajectory coefficients at leading order in $G$ and at every logarithmic order using Legendre
polynomials. The first two coefficients reproduce the known results at orders $G$ and $G^2$.
Read the full abstract ▾
Artificial Intelligence
AGAR: a reinforcement learning substrate for LLM program evolution
Haoran Li, Zengle Ge, Xiaomin Yuan et al. (+11 more)
Given a task and an evaluator, a
language model can rewrite a candidate program while a search loop decides which rewrites survive, offering a practical route to
algorithm discovery. But that loop is governed by five constants set by hand: which parent to select, how hard to mutate, how to keep diversity, what to remember, and a scalar score that never says which part of the program earned it. Reinforcement learning already has an
estimator for each. The obstacle is that program
evolution is not usually written down as a decision process. We formalize it as a Markov decision process whose action is the modular prefix the model is conditioned on, rather than the program it emits. Credit assignment, value estimation, adaptive exploration, and experience
memory can then attach to distinct components. AGAR (
Algorithm Generation As RL) provides the resulting substrate: any
estimator can be replaced or switched off without changing the controller, making the transfer auditable one mechanism at a time, with no gradient training of the backend model. Across 19 tasks, two backends, and three seeds under one harness, AGAR improves on the stronger of two published baselines on most tasks, with gains concentrated in the competitive-programming family. The formalization also yields a checkable reading of prior work: these systems are implicitly zero-discount, not by choice, but because fitness is exogenous to an individual rather than a return over successors, leaving a discount factor nothing to act on.
Read the full abstract ▾
All authors Haoran Li, Zengle Ge, Xiaomin Yuan, Yui Lo, Haoxin Li, Songlin Zhou, Qianhui Liu, Jiahua Ying, Yuanhang Liu, Mingju Chen, Annan Li, Jianmin Wu, Dawei Yin, Dou Shen
Statistical Methodology
Calibration-Free Robust Bayesian Inference for Discrete Unnormalized Models
Jiongran Wang, Debdeep Pati and Anirban Bhattacharya
Bayesian inference for discrete models with intractable normalizing constants is computationally challenging, particularly when the sample space is large or countably infinite. Although some likelihood-based
Bayesian methods can circumvent the intractable normalizing constant, their uncertainty quantification may be poorly calibrated under model misspecification. Generalized
Bayesian approaches can address this calibration issue, but typically require careful tuning of a learning rate. In this paper, we propose DFD-BETEL, a semiparametric
Bayesian framework that uses the first-order optimality conditions of discrete Fisher divergence as moment conditions in
Bayesian exponentially tilted empirical likelihood. The resulting posterior avoids evaluation of the normalizing constant and requires no learning-rate calibration. We establish consistency and asymptotic normality of the discrete Fisher divergence
estimator and prove a Bernstein-von Mises
theorem for the DFD-BETEL posterior, showing that its limiting covariance matches the sampling covariance of the
estimator and yields asymptotically calibrated uncertainty quantification. Numerical studies show that DFD-BETEL remains competitive with likelihood-based
Bayesian inference under correct specification and provides more reliable uncertainty quantification under model misspecification. An application to automobile-insurance claim data further illustrates the practical utility of DFD-BETEL.
Read the full abstract ▾
Systems and Control
Mission-critical spectrum sharing with decentralized Multi-Agent Reinforcement Learning
Dimitrios Pylorof, Imtiaz Nasim, Humberto E. Garcia et al. (+3 more)
Motivated by emerging mission-critical applications and an increasingly congested spectrum, we develop a decentralized multi-
agent reinforcement learning (MARL) model for dynamic spectrum access. The model enables secondary users to learn effective transmission strategies across shared frequency bands while minimizing collisions with high-priority primary users and among themselves. We design the
agent-level learners following a Markov potential game approach, connecting independent local updates to system-level improvement. We instantiate this design using lightweight linear actor-critic learners suitable for resource-constrained edge devices, rather than computationally intensive centralized or deep multi-
agent architectures. Across spectrum environments with different incumbent activities, the learned policies adapt their transmission policy and waiting behavior to preserve throughput while greatly reducing transmission collisions relative to random and forecast-aware heuristic baselines. The results establish the value of decentralized MARL and shows up to 96.8% reduction in overall collisions.
Read the full abstract ▾
All authors Dimitrios Pylorof, Imtiaz Nasim, Humberto E. Garcia, Vivek Agarwal, Jasni A. Mannil, Mingyue Ji
Machine Learning
CurveTQ: Rotation-Free Trellis Quantization of LLM Weights via Curvature-Weighted Search
Guanhua Ding, Zi Wang, Ruichao Li et al. (+1 more)
The best two-bit weight quantizers for large
language models, such as QTIP and Proteus, rotate each weight
matrix by a random orthogonal transform, which must be undone at every decoding step, then encode it with a trellis or
lattice code under a Euclidean search; the layer Hessian enters only through error feedback between coding blocks. We show that this leaves part of the Hessian unused. Error feedback turns the loss into a weighted sum of per-coordinate rounding errors whose weights, the diagonal of the Hessian's LDL factorization, existing quantizers compute but never read. We put these weights into the Viterbi branch metric, so the search follows the
curvature within each coding block. This also explains the rotation: it removes this within-block variation, so weighting in the native basis and rotating are substitutes. On three models the weighted native search matches a full-dimension randomized Hadamard to within about one point of downstream accuracy, and weighting after the rotation gains little. Around this search we build CurveTQ, a trellis codec with no rotation, which handles the weights' amplitude and marginal shape with a factored scale field and a closed-form
quantile table, and stores a start state per coding block so the trellis can adapt to the residual that error feedback carries into it. At two bits CurveTQ is 1-3 points higher in mean downstream accuracy than QTIP and Proteus on three 4-8B Instruct models, even after both are given our start state, which alone lifts either baseline by 1-3 points. It also leads on a 35B mixture of experts, to our knowledge the first trellis-coded result on such a model. With no rotation to undo, our decoder is the fastest of the three at all tested batch sizes and bit widths.
Read the full abstract ▾
All authors Guanhua Ding, Zi Wang, Ruichao Li, Jack Liu
Sound
BanglaBox: A Phonetically-Balanced Corpus and Data-Efficient Foundation-Model Adaptation for Bangla Text-to-Speech with Zero-Shot Voice Cloning
Emtiaz Uddin Ahmed, Araf Mahmud, Sajib Hossain et al. (+3 more)
We present a recipe for adapting English-pretrained autoregressive TTS foundation models to underrepresented
languages, demonstrated on Bangladeshi Bangla. Existing Bangla TTS corpora are small and single-speaker, and to our knowledge no open zero-shot voice-cloning system is available for the Bangladeshi register. We contribute a phonetically- and gender-balanced two-tier Bangladeshi Bangla corpus balanced via a tiered Jensen-Shannon divergence objective over conjunct clusters (juktakkhor), together with three fine-tuning changes: a merge-consistent tokenizer extension, Bangla text normalization, and a
prompt-masked dual-loss objective. These changes preserve zero-shot cloning across the
language switch. Our BanglaEval protocol applies Wilcoxon signed-rank tests with Bonferroni correction over native-speaker ratings. BanglaBox attains near-natural Naturalness, outperforms commercial and open-source baselines on Naturalness, Speaker Similarity, and Clarity, and reaches speaker similarity comparable to prior few-shot results while using approximately 7x less Bangla fine-tuning audio. We further validate the recipe beyond our own test split using a seven-category stress set of difficult real-world text, the public BnTTS evaluation
benchmarks, and naturally occurring Bangla that no
language model wrote. All artifacts, including the corpus, weights, code, and complete evaluation materials, are released publicly and unconditionally.
Read the full abstract ▾
All authors Emtiaz Uddin Ahmed, Araf Mahmud, Sajib Hossain, Tarikul Islam Tamiti, Sajid Fardin Dipto, Anomadarshi Barua
Strongly Correlated Electrons
Multiorbital Anisotropy and Magnetic Excitations in Trilayer Nickelate Pr$_4$Ni$_3$O$_{10}$
Ziqiang Guan, Sophia F. R. TenHuisen, Xuefei Guo et al. (+9 more)
Rare-earth substitution strongly modifies the structure,
metallicity, and superconducting behavior of Ruddlesden-Popper (RP) nickelates, but its impact on their
magnetic interactions remains unclear. Here, we combine x-ray absorption spectroscopy (XAS) and resonant inelastic x-ray scattering (RIXS) to resolve the multiorbital electronic structure and
magnetic excitations of Pr$_4$Ni$_3$O$_{10}$. The absence of an overlapping rare-earth resonance provides direct access to the Ni $L_3$ edge, while polarization-dependent measurements at the Ni $L$- and O $K$-edges resolve distinct planar and out-of-plane Ni-O hybridization channels. Momentum-resolved RIXS reveals dispersive
magnetic excitations that soften toward the incommensurate
spin-density-wave wave vector. An effective
spin-wave analysis yields a sizable intra-trilayer exchange coupling, comparable to the leading in-plane interactions and robust against rare-earth substitution. Our results identify the multilayer architecture, rather than rare-earth substitution alone, as a key control knob of the
magnetic interactions in these materials.
Read the full abstract ▾
All authors Ziqiang Guan, Sophia F. R. TenHuisen, Xuefei Guo, Wei He, Aulden K. Jones, Marli dos Reis Cantarino, Xinglong Chen, Hong Zheng, J. F. Mitchell, Nicholas B. Brookes, Mark P. M. Dean, Matteo Mitrano
Computation and Language
Multi-Objective Aligned Small Language Model Framework for SUD Patient Dialogue Generation
Thushara Manjari Naduvilakandy, Hyeju Jang and Mohammad Al Hasan
Substance Use Disorder (SUD) counseling requires patient responses that reflect underlying cognitive states such as beliefs, coping strategies, and readiness for change. Although large
language models (LLMs) can generate fluent text, they often fail to produce cognitively coherent and clinically realistic patient behavior, especially under ethical and data-scarce clinical settings. Moreover, deploying frontier-scale LLMs in healthcare applications presents practical challenges including high computational cost, latency,
privacy concerns, and limited deployability in resource-constrained environments, motivating the need for cognitively aligned small
language models (SLMs). We propose a cognitively grounded framework for SUD patient dialogue generation that explicitly models and aligns latent cognitive components with patient histories and counselor questions. Our pipeline consists of two stages: cognitive component detection and cognitive component-aligned dialogue generation. To enable effective learning with smaller models, we combine knowledge
distillation from high-capacity teacher models, preference optimization from human-annotations, and
attention-guided reward shaping. Extensive evaluations using automatic scores like BERTScore, ROUGE, METEOR and BLEU, and LLM-as-judge hit-metrics against both human and teacher-model references show that cognitively informed fine-tuning substantially improves cognitive realization and
alignment over a generic instruction-tuned baselines and mental health domain specific SLMs, with particularly strong gains for open-ended cognitive components.
Read the full abstract ▾
Statistical Methodology
Sketched Calibration for Conformal Prediction under Covariate Shift
Mehrdad Pournaderi
Weighted conformal prediction corrects for
covariate shift by reweighting calibration scores with the likelihood ratio between target and source
covariates. Its cost grows with the chi-square divergence between the two
covariate laws, typically exponentially in the size of the shift, and much of it can be paid for shift in directions that do not affect the response. We propose sketched calibration: weighted conformal prediction with the ratio of compressed
covariates $Z=T(X)$, used in the weights and optionally in the score. Compression never increases the shift-dependent calibration cost, and target coverage is at least $1-\alpha-\Delta_T$, where the leakage $\Delta_T$ measures how much of the discarded shift reappears as a change in the law of the response, or of the score, given $Z$. The leakage is a covariance between the discarded shift and the response on the fibres of the sketch; it vanishes when the sketch is sufficient or retains the shift, and a minimax construction shows that no threshold rule for a fixed score can avoid it. In a heteroscedastic simulation where unweighted calibration fails, a one-dimensional sketch cuts the fraction of infinite prediction sets from $37.8\%$ to $8.0\%$ at mean coverage $92.3\%$ for a $90\%$ target; nonlinear and real-data experiments show when learned sketches succeed and when they fail.
Read the full abstract ▾
Robotics
Belief-Space Planning with Planner-Conditioned Estimator Error under Intermittent Observations
Peter M. Donley and Joseph L. Moore
Belief-space planners with separately designed or off-the-shelf
estimators may have access to state-relevant information the
estimator does not observe. Consequently, even an
estimator that minimizes mean-squared error under its own information can have a nonzero error mean when conditioned on planner information. When future corrections are intermittent and
stochastic, differences between accepted and rejected error means introduce additional uncertainty terms to correction events that zero-mean assumptions ignore. In this paper, we propose a belief-space planning approach that predicts, propagates, and penalizes planner-conditioned
estimator error along evaluated trajectories. A planner-conditioned
estimator-error model combines affine error dynamics with a moment recursion that preserves stochastically-induced correction uncertainty. We describe a method by which to predict
estimator error dynamics within specified operating regimes and incorporate predicted error moments into a task-weighted quadratic risk objective. We evaluate the approach in simulation for a tilt-rotor VTOL landing on a ship deck using receding-horizon planning. We find that conditioning on planner information reduced error-prediction loss for a command-blind EKF by 13.4% relative to a planner-ignorant model, with strongly regime and horizon-dependent
forecasting abilities. In a small set of 16 paired closed-loop trials, the planner lowered the median terminal task gauge from 1.60 to 0.99, and predicted
estimator error beyond 2s within a factor of 1.3, versus a 5.7-fold underprediction by covariance-only planning.
Read the full abstract ▾
Machine Learning
An Accuracy-Information Tradeoff for Loss-Difference Conditional Mutual Information
Hazar Yueksel
Loss-difference conditional mutual information (ld-CMI) uses the smallest of the standard observations in the supersample hierarchy of generalization bounds: it measures what a learner's loss differences reveal about which candidate of each pair it was trained on. Accuracy is known to force information into the model; data processing does not carry such lower bounds to losses. We show, by bounding three moments of the loss differences, that accuracy also forces ld-CMI. For linear predictors with a smooth convex loss of nonzero slope at zero, such as the logistic loss, plus a regularizer whose
curvature and growth are both of power $r\ge2$, on product distributions over a scaled sign cube in dimension at least linear in $n$, every proper learner with expected excess risk at most $\varepsilon$ on these distributions at the optimal sample size $n\asymp\varepsilon^{-2+2/r}$ has worst-case ld-CMI of order $n$ bits, and $\Theta(n/(1+(\tau/\varepsilon)^2))$ bits under Gaussian noise of standard deviation $\tau$ on the loss differences. The same holds without a regularizer, at $n\asymp\varepsilon^{-2}$. Consequently, range-scaled ld-CMI bounds cannot vanish on these distributions, although every proper learner's generalization gap is $O(n^{-1/2})$. We also show that model-level information does not determine noisy loss-difference information, and that the growth, slope and dimension conditions are needed, the last up to a logarithm.
Read the full abstract ▾
Computer Vision
Do Vision Models Learn Physical Constraints or Rendering Shortcuts? A Counterfactual Benchmark for Grounded Physical Consistency
M. Moein Esfahani, Sepehr Salem, Mohammed Alser et al. (+1 more)
Modern image editing models can satisfy a text instruction while breaking the physics of the edited scene. A new object may cast no shadow, a mirror may fail to reflect visible geometry, or an object may float above a surface that should support it. We study physical plausibility diagnosis, detecting whether an edited image violates scene physics, naming the violation type, localizing the affected region, and explaining the failure in
language. We introduce a counterfactual
benchmark whose controlled synthetic component uses Mitsuba~3 to generate 5,500 images from 500 scene families. Each family contains one clean image and ten matched violations involving shadows, reflection, support, surface response, and occlusion. The renderer pipeline provides category labels, affected-region masks and boxes, scene metadata, and explanation targets. We use LLaVA-1.5-7B, Qwen2.5-VL-7B, and InternVL3.5-8B as diagnostic baselines rather than proposed methods. On a 1,650-image synthetic test set, the adapted baselines reach 64.0–67.8% category macro-F1 on standard held-out scenes. For LLaVA-1.5-7B, category macro-F1 falls from 64.0% on the standard split to 40.8% under intervention shift. This gap shows that high in-distribution accuracy partly reflects cues tied to rendering and counterfactual construction.
Read the full abstract ▾
All authors M. Moein Esfahani, Sepehr Salem, Mohammed Alser, Vince Calhoun
Soft Condensed Matter
Contamination dynamics of active rods in long microchannels
Jason Palos, Chase Brown, Mykhailo Potomkin et al. (+1 more)
Active rods are elongated, self-propelled particles that serve as effective models for motile microorganisms, such as bacteria and sperm
cells. When confined to a microfluidic channel, active rods exhibit a distinctive motility pattern: they tend to swim against the fluid flow along channel walls. In pressure-driven channel flow, they alternate between long upstream runs along the walls and short downstream spurts in the bulk. This behavior has practical implications for bacterial contamination of catheters and microfluidic devices against the flow. We study the spreading of a population of active rods along long microchannels using a hierarchy of models. Monte Carlo simulations of a three-dimensional model show that net upstream transport depends non-monotonically on the flow rate, decreases with rotational
diffusion, and depends on the shape of the channel cross-section. The simulations also show that upstream and downstream persistence times follow exponentially decaying distributions. This motivates the derivation of a one-dimensional, two-state velocity-jump model. For exponentially distributed persistence times, we derive the
evolution equation for the concentration of continuously injected rods and determine its long-time limit in closed form. This analysis reveals a concentration plateau upstream of the source when the mean upstream velocity is positive and exponential decay otherwise. Using parameters measured from the three-dimensional simulations, the one-dimensional model reproduces the concentration profiles from the Monte Carlo simulations, and these results are consistent with recent experimental observations of bacterial contamination in microfluidic devices.
Read the full abstract ▾
All authors Jason Palos, Chase Brown, Mykhailo Potomkin, Shawn D. Ryan
Optimization and Control
FlexGrad: a root-free approach to adaptive step sizes
Pablo Barros, Aaron Defazio and Vincent Guigues
We introduce the FlexGrad optimizer, an adaptive gradient method that addresses a key limitation of classical AdaGrad-type methods: their inability to exploit correlations between gradients. By scaling steps using both cumulative gradients and cumulative squared gradients, FlexGrad preserves AdaGrad-like behavior in noisy regimes while allowing substantially larger steps when gradient directions are persistent. The method is fully online, horizon-free, and avoids any auxiliary search procedure during optimization. For convex Lipschitz objectives, we prove that FlexGrad attains the optimal asymptotic convergence rate for nonsmooth convex optimization, with guarantees for both global and coordinate-wise variants. FlexGrad is straightforward to implement and is closely connected to practical gradient-normalized methods. Experiments on a range of convex and nonconvex problems show that FlexGrad performs competitively across a range of optimization settings.
Read the full abstract ▾
Artificial Intelligence
Few Bits, One Law: Toward W2A4KV2
Kai Yi, Tarek Elgamal, Sruthikesh Surineni et al. (+3 more)
Extreme low-bit LLM compression is most challenging when weights, activations, and KV caches are quantized together: their distributions differ, and quantization errors interact throughout the network. We introduce CanonQ, a unified quantization-aware training framework that addresses these challenges by separating source canonicalization from task-aware adaptation. Fixed rotations and energy normalization map heterogeneous tensor sources to canonical coordinates, enabling frozen Gaussian-reference codebooks to be reused across layers and models. Joint training then adapts the network to the coupled errors of weight, activation, and cache quantization within a common scalar/vector interface. We bound frozen-codebook transfer error and local task loss, and derive an exact normalization-aware straight-through Jacobian that links quantization distortion to gradient bias. The strongest gains arise under joint W2A4KV2 compression: across LLaMA3-1B/3B/8B, CanonQ-Omni achieves up to 14.28x lower WikiText-2 perplexity and up to 57.9% higher mean zero-shot accuracy than prior state-of-the-art and representative quantization baselines. The benefits extend to Qwen3-1.7B, code generation, and mathematical
reasoning: on instruction-tuned MobileLLM-Pro-1B at W2A16KV16, CanonQ achieves relative improvements of 41.7% in HumanEval pass@1 and 39.1% in GSM8K exact match over the strongest evaluated quantization baseline.
Read the full abstract ▾
All authors Kai Yi, Tarek Elgamal, Sruthikesh Surineni, Vignesh Vivekraja, Soumyadeep Ghosh, Steven Li
Geometric Topology
Cabling formula and colored knot Floer homology
Chen Zhang
We study colored
knot Floer homology via a bordered cabling construction. We construct a cabling module and identify the connecting maps in the defining direct system at the level of this module. By taking the colimit at the chain level, we obtain a general formula for colored
knot Floer homology of an arbitrary
knot in terms of its ordinary
knot Floer homology and the colored unknot
algebra. This extends the computation of Alishahi-Gorsky-Liu from L-space
knots to all
knots.
Read the full abstract ▾
Computer Vision
StyleFields: Multi-Scale AdaIN-Modulated Implicit SDFs for Coarse-to-Fine 3D Shape Reconstruction and Editing
Ehsan Garaaghaji, Nicolas Talabot, Pascal Fua et al. (+1 more)
We introduce StyleFields, a DeepSDF-based architecture for high-fidelity 3D reconstruction that enables controllable
geometric style mixing: the coarse structure of one object can be combined with the fine-scale details of another. The core idea is depth-aware modulation: instead of a single global code, we inject latents via multi-level Adaptive Instance Normalization at several decoder depths, and supervise matching auxiliary heads with a coarse-to-fine schedule while gradually growing network depth. This aligns early layers with global shape and later layers with high-frequency detail, achieving content-style decoupling without part labels or
adversarial training. StyleFields delivers faithful reconstructions, convincing cross-instance hybrids, and consistent gains in ablations over injection depth and supervision granularity. We further demonstrate a practical application in automotive aerodynamics: a learned surrogate drag predictor serves as a differentiable objective to optimize reconstructed cars, allowing targeted edits of global form or surface details by freezing the complementary latent stream. StyleFields offers a simple, effective recipe for controllable implicit reconstruction and downstream performance-driven design.
Read the full abstract ▾
All authors Ehsan Garaaghaji, Nicolas Talabot, Pascal Fua, Doruk Oner
General Relativity and Quantum Cosmology
Toward a complete understanding of the properties of GW190521
Aasim Jan, Sophia Nicolella, Aaron Zimmerman et al. (+4 more)
Interpreting GW190521 remains challenging due to its exceptionally high total mass and correspondingly small number of observable gravitational-wave cycles, which gives rise to degeneracies between different physical interpretations. Motivated by the outstanding questions surrounding the event, we present a comprehensive reanalysis of GW190521 using waveform models drawn from multiple families, including a model that includes both eccentricity and
spin-precession. We find evidence that the source was highly eccentric, with an inferred eccentricity of $0.81^{+0.05}_{-0.01}$ at 5 Hz, and show that non-eccentric interpretations arise from failing to probe the high-likelihood region at large eccentricities. We further demonstrate that when eccentricity is included in the analysis, there is no indication of
spin precession, suggesting that the apparent precession in quasicircular analyses is at least partially driven by an eccentricity–
spin-precession degeneracy. While all waveform models favor a comparable-mass binary, one model admits a secondary mode at more asymmetric masses. We find a robust probability $(\geq97\%)$ that the primary black hole lies within the pair-instability
supernova mass gap ($\sim60-130\,M_\odot$), whereas the probability that the secondary lies in the gap depends on the waveform model. Through inference of GW190521-like injections, we show that the waveform systematics observed in the analysis of GW190521 could arise from the response of models to unmodeled eccentricity, rather than from intrinsic differences between waveform models. When analyzing a quasicircular
spin-precessing injection similar to GW190521, we find no significant model dependence in the inferred parameters. Finally, because the inferred eccentricity lies beyond the calibration regime of current models, we test its robustness by recovering highly eccentric numerical-relativity waveforms.
Read the full abstract ▾
All authors Aasim Jan, Sophia Nicolella, Aaron Zimmerman, Deirdre Shoemaker, Katerina Chatziioannou, Richard O'Shaughnessy, Muhammed Saleem
Programming Languages
Modular Responsiveness Verification of Rust Async Runtimes
Yanze Li, Ivan Beschastnikh and Alexander J. Summers
Asynchronous (async) programming is a popular paradigm for managing concurrency.
Languages that provide async support typically have a runtime to manage asynchronous executions. These runtimes are critical infrastructure, yet verifying them has received little
attention. One reason is that a property that users care most about from an async runtime is a liveness property: tasks submitted to the runtime eventually make progress. Verifying liveness is challenging for libraries that are both concurrent and highly optimized. We present a lightweight and modular proof technique for verifying eventual progression guarantees for Rust async runtimes. We describe this technique in the context of a simple
language based on Rust and Rust's async model. We then realize this proof technique as a set of static analyses for Rust and use these to verify eventual progression of several key components of multiple Rust async runtime implementations.
Read the full abstract ▾
Artificial Intelligence
FreeEvolve: Learning to Evolve Beyond Fixed Loops
Lecheng Kong, Like Hui, Nikos Kanakaris et al. (+3 more)
Agent evolvers automate the design of the
prompts, skills and workflows around
language model
agents, yet the optimization process they follow is still designed by hand: a fixed search loop decides how candidates are evaluated, which are kept and when the search stops. We propose FREEEVOLVE, which automates this process as well. An environment specifies the goal, target
agent, evaluator, data and resource limits; within these limits, the evolver itself decides what to test, how much evidence to collect, which candidates to pursue and when to stop. These decisions follow an editable
evolution skill, which we improve through meta-
evolution by scoring each candidate skill on the fresh target
agent it produces. The optimization process thus becomes a capability learned from experience rather than a loop engineered in advance. On tau3-bench, ARC-AGI-2, ARC-AGI-3 and Terminal-Bench 2.1, FREEEVOLVE controls the
evolution campaign by itself, yet improves the primary held-out metric by 13.6 points on average and matches or exceeds hand-designed evolvers. The learned process keeps improving with experience: meta-evolved skills add 6.9 points over the seed skill on fresh target
agents, demonstrating transferability across environments.
Read the full abstract ▾
All authors Lecheng Kong, Like Hui, Nikos Kanakaris, Prithwish Jana, Sahika Genc, Narayanan Sadagopan
Machine Learning
Exact Dynamics and Finite-Sample Trajectory Recovery of Linear Recursive Feature Machines
Andrew Cheng, Bobak T. Kiani, Yue M. Lu et al. (+1 more)
Recursive feature machines (RFMs) learn representations of data by alternating between fitting a predictor to a
dataset and updating features of that predictor using the average gradient outer product (AGOP). Connections between AGOPs and feature learning in neural networks motivate linear RFMs as a simple setting for analyzing how representations evolve during training. Here, we study the dynamics and statistics of linear RFM in noisy multi-output regression with isotropic sub-Gaussian input data and targets generated by a low-rank teacher
matrix of dimension $d$. We extend the known connection between linear RFM and iteratively reweighted least squares from the interpolating setting to ridge-regularized multi-output regression with noise. We show that the learned feature
matrix remains close to its infinite-data ideal counterpart at every iteration. Namely, for $n$ samples, we show the error in the feature
matrix decays as $O(\sqrt{d/n})$ with high probability. Experiments on real-world text and single-
cell gene-expression data illustrate the features learned by this simple linear model.
Read the full abstract ▾
All authors Andrew Cheng, Bobak T. Kiani, Yue M. Lu, Adityanarayanan Radhakrishnan
Computer Vision
StableGrasp: Reconstructing Physically Stable Human Hand Grasps from Single Images
Han Jiang, Etienne Vouga, Qixing Huang et al. (+1 more)
Reconstructing a physically stable human grasp from a single RGB image is challenging because physically modeling grasps is itself difficult, and the problem requires estimating not only a visually constrained hand pose but also a control target that stabilizes the grasp. Existing methods either model only visual hand geometry without considering physics, or rely on less plausible physical modeling, which limits the physical validity of the resulting grasps. In this paper, we present StableGrasp, a differentiable simulation-based optimization framework that explicitly separates the visual hand pose from the control target that determines the grasping forces. Our method jointly optimizes hand geometry and control by minimizing the kinetic energy of the grasp in a differentiable simulator, while regularizing the hand geometry to preserve visual consistency and
geometric plausibility. The reconstructed grasps are substantially more stable under rigorous physical simulation, while remaining visually consistent with the input images and geometrically plausible. Experiments show that our approach produces far more stable grasps than alternative hand-control strategies, benefiting visual-only grasp reconstruction pipelines by turning their outputs into physically stable grasps.
Read the full abstract ▾
All authors Han Jiang, Etienne Vouga, Qixing Huang, Georgios Pavlakos