id
stringlengths
9
16
title
stringlengths
4
278
categories
stringlengths
5
104
abstract
stringlengths
6
4.09k
2502.04530
Robust Probabilistic Model Checking with Continuous Reward Domains
cs.AI cs.FL cs.LG
Probabilistic model checking traditionally verifies properties on the expected value of a measure of interest. This restriction may fail to capture the quality of service of a significant proportion of a system's runs, especially when the probability distribution of the measure of interest is poorly represented by its expected value due to heavy-tail behaviors or multiple modalities. Recent works inspired by distributional reinforcement learning use discrete histograms to approximate integer reward distribution, but they struggle with continuous reward space and present challenges in balancing accuracy and scalability. We propose a novel method for handling both continuous and discrete reward distributions in Discrete Time Markov Chains using moment matching with Erlang mixtures. By analytically deriving higher-order moments through Moment Generating Functions, our method approximates the reward distribution with theoretically bounded error while preserving the statistical properties of the true distribution. This detailed distributional insight enables the formulation and robust model checking of quality properties based on the entire reward distribution function, rather than restricting to its expected value. We include a theoretical foundation ensuring bounded approximation errors, along with an experimental evaluation demonstrating our method's accuracy and scalability in practical model-checking problems.
2502.04531
AnyPlace: Learning Generalized Object Placement for Robot Manipulation
cs.RO cs.AI cs.CV
Object placement in robotic tasks is inherently challenging due to the diversity of object geometries and placement configurations. To address this, we propose AnyPlace, a two-stage method trained entirely on synthetic data, capable of predicting a wide range of feasible placement poses for real-world tasks. Our key insight is that by leveraging a Vision-Language Model (VLM) to identify rough placement locations, we focus only on the relevant regions for local placement, which enables us to train the low-level placement-pose-prediction model to capture diverse placements efficiently. For training, we generate a fully synthetic dataset of randomly generated objects in different placement configurations (insertion, stacking, hanging) and train local placement-prediction models. We conduct extensive evaluations in simulation, demonstrating that our method outperforms baselines in terms of success rate, coverage of possible placement modes, and precision. In real-world experiments, we show how our approach directly transfers models trained purely on synthetic data to the real world, where it successfully performs placements in scenarios where other models struggle -- such as with varying object geometries, diverse placement modes, and achieving high precision for fine placement. More at: https://any-place.github.io.
2502.04533
Polarization-Dependent Loss Mitigation with Orthogonal-Design-Based Precoding and Interference Cancellation
cs.IT math.IT
Recent work by Shehadeh and Kschischang provides a simple capacity-achieving scheme for channels with polarization-dependent loss (PDL) under common modeling assumptions via a careful choice of orthogonal-design-based precoding and interference cancellation. This letter extends that work with a simulation-based demonstration showing that this scheme remains highly effective at mitigating PDL in the highly practical setting of 4-PAM with Chase-decoded extended Hamming inner codes rather than the near-capacity inner codes considered in the original work. An alternative near-optimal variation of this scheme is provided requiring only one inner code rather than two and suffering no penalty in the absence of PDL, making it much more practical.
2502.04535
A Decoding Algorithm for Length-Control Summarization Based on Directed Acyclic Transformers
cs.CL
Length-control summarization aims to condense long texts into a short one within a certain length limit. Previous approaches often use autoregressive (AR) models and treat the length requirement as a soft constraint, which may not always be satisfied. In this study, we propose a novel length-control decoding algorithm based on the Directed Acyclic Transformer (DAT). Our approach allows for multiple plausible sequence fragments and predicts a \emph{path} to connect them. In addition, we propose a Sequence Maximum a Posteriori (SeqMAP) decoding algorithm that marginalizes different possible paths and finds the most probable summary satisfying the length budget. Our algorithm is based on beam search, which further facilitates a reranker for performance improvement. Experimental results on the Gigaword and DUC2004 datasets demonstrate our state-of-the-art performance for length-control summarization.
2502.04537
Multilingual Non-Autoregressive Machine Translation without Knowledge Distillation
cs.CL
Multilingual neural machine translation (MNMT) aims at using one single model for multiple translation directions. Recent work applies non-autoregressive Transformers to improve the efficiency of MNMT, but requires expensive knowledge distillation (KD) processes. To this end, we propose an M-DAT approach to non-autoregressive multilingual machine translation. Our system leverages the recent advance of the directed acyclic Transformer (DAT), which does not require KD. We further propose a pivot back-translation (PivotBT) approach to improve the generalization to unseen translation directions. Experiments show that our M-DAT achieves state-of-the-art performance in non-autoregressive MNMT.
2502.04541
The Phantom of the Elytra -- Phylogenetic Trait Extraction from Images of Rove Beetles Using Deep Learning -- Is the Mask Enough?
cs.CV
Phylogenetic analysis traditionally relies on labor-intensive manual extraction of morphological traits, limiting its scalability for large datasets. Recent advances in deep learning offer the potential to automate this process, but the effectiveness of different morphological representations for phylogenetic trait extraction remains poorly understood. In this study, we compare the performance of deep learning models using three distinct morphological representations - full segmentations, binary masks, and Fourier descriptors of beetle outlines. We test this on the Rove-Tree-11 dataset, a curated collection of images from 215 rove beetle species. Our results demonstrate that the mask-based model outperformed the others, achieving a normalized Align Score of 0.33 plus/minus 0.02 on the test set, compared to 0.45 plus/minus 0.01 for the Fourier-based model and 0.39 plus/minus 0.07 for the segmentation-based model. The performance of the mask-based model likely reflects its ability to capture shape features while taking advantage of the depth and capacity of the ResNet50 architecture. These results also indicate that dorsal textural features, at least in this group of beetles, may be of lowered phylogenetic relevance, though further investigation is necessary to confirm this. In contrast, the Fourier-based model suffered from reduced capacity and occasional inaccuracies in outline approximations, particularly in fine structures like legs. These findings highlight the importance of selecting appropriate morphological representations for automated phylogenetic studies and the need for further research into explainability in automatic morphological trait extraction.
2502.04543
Sparsity-Based Interpolation of External, Internal and Swap Regret
stat.ML cs.LG
Focusing on the expert problem in online learning, this paper studies the interpolation of several performance metrics via $\phi$-regret minimization, which measures the performance of an algorithm by its regret with respect to an arbitrary action modification rule $\phi$. With $d$ experts and $T\gg d$ rounds in total, we present a single algorithm achieving the instance-adaptive $\phi$-regret bound \begin{equation*} \tilde O\left(\min\left\{\sqrt{d-d^{\mathrm{unif}}_\phi+1},\sqrt{d-d^{\mathrm{self}}_\phi}\right\}\cdot\sqrt{T}\right), \end{equation*} where $d^{\mathrm{unif}}_\phi$ is the maximum amount of experts modified identically by $\phi$, and $d^{\mathrm{self}}_\phi$ is the amount of experts that $\phi$ trivially modifies to themselves. By recovering the optimal $O(\sqrt{T\log d})$ external regret bound when $d^{\mathrm{unif}}_\phi=d$, the standard $\tilde O(\sqrt{T})$ internal regret bound when $d^{\mathrm{self}}_\phi=d-1$ and the optimal $\tilde O(\sqrt{dT})$ swap regret bound in the worst case, we improve existing results in the intermediate regimes. In addition, the same algorithm achieves the optimal quantile regret bound, which corresponds to even easier settings of $\phi$ than the external regret. Building on the classical reduction from $\phi$-regret minimization to external regret minimization on stochastic matrices, our main idea is to further convert the latter to online linear regression using Haar-wavelet-inspired matrix features. Then, we apply a particular $L_1$-version of comparator-adaptive online learning algorithms to exploit the sparsity in this regression subroutine.
2502.04544
Solvability of Approximate Reach-Avoid Games
eess.SY cs.LO cs.SY
Objective: In a companion paper, we propose a parametric hybrid automaton model and an algorithm for the online synthesis of robustly correct and near-optimal controllers for cyber-physical system with reach-avoid guarantees. A key part of this synthesis problem is based on a weighted discretised game and solved via scope-adaptive discrete dynamic programming. Approach: This work examines proofs of key properties of the discussed algorithm. Evaluation: The main proof is by induction over the stages of a discrete Hamilton-Jacobi-Bellman system of equations. Contribution: The results include a game solvability theorem and identify necessary and sufficient conditions for its applicability.
2502.04548
Contextual Gradient Flow Modeling for Large Language Model Generalization in Multi-Scale Feature Spaces
cs.CL
Optimization methodologies for training large-scale neural architectures often rely on uniform gradient propagation mechanisms that fail to align with hierarchical linguistic structures, limiting their capacity to generalize across diverse language distributions. A structured gradient refinement framework was introduced to incorporate multi-scale contextual adjustments, improving parameter adaptation through dynamic weighting strategies that enhanced representation coherence. Empirical evaluations demonstrated that structured propagation mechanisms contributed to reductions in gradient oscillations, resulting in more stable training dynamics and improved optimization efficiency. The comparative performance assessment indicated that models incorporating hierarchical propagation strategies exhibited greater robustness in long-range dependency retention and cross-domain adaptation. The hierarchical adjustment of weight updates provided an alternative to conventional backpropagation, reducing sensitivity to initialization conditions while improving overall convergence efficiency. The experimental results confirmed that structured gradient propagation influenced representation learning trajectories, aligning parameter updates with broader linguistic dependencies rather than isolated token-level relationships. Statistical evaluations indicated that structured optimization strategies mitigated overfitting while preserving adaptability across heterogeneous text distributions. The findings established that structured gradient propagation provided an empirically validated framework for refining hierarchical representation learning, supporting more effective integration of linguistic dependencies into optimization dynamics.
2502.04549
Mechanisms of Projective Composition of Diffusion Models
cs.LG
We study the theoretical foundations of composition in diffusion models, with a particular focus on out-of-distribution extrapolation and length-generalization. Prior work has shown that composing distributions via linear score combination can achieve promising results, including length-generalization in some cases (Du et al., 2023; Liu et al., 2022). However, our theoretical understanding of how and why such compositions work remains incomplete. In fact, it is not even entirely clear what it means for composition to "work". This paper starts to address these fundamental gaps. We begin by precisely defining one possible desired result of composition, which we call projective composition. Then, we investigate: (1) when linear score combinations provably achieve projective composition, (2) whether reverse-diffusion sampling can generate the desired composition, and (3) the conditions under which composition fails. Finally, we connect our theoretical analysis to prior empirical observations where composition has either worked or failed, for reasons that were unclear at the time.
2502.04552
Reinforcement Learning Based Prediction of PID Controller Gains for Quadrotor UAVs
eess.SY cs.RO cs.SY
A reinforcement learning (RL) based methodology is proposed and implemented for online fine-tuning of PID controller gains, thus, improving quadrotor effective and accurate trajectory tracking. The RL agent is first trained offline on a quadrotor PID attitude controller and then validated through simulations and experimental flights. RL exploits a Deep Deterministic Policy Gradient (DDPG) algorithm, which is an off-policy actor-critic method. Training and simulation studies are performed using Matlab/Simulink and the UAV Toolbox Support Package for PX4 Autopilots. Performance evaluation and comparison studies are performed between the hand-tuned and RL-based tuned approaches. The results show that the controller parameters based on RL are adjusted during flights, achieving the smallest attitude errors, thus significantly improving attitude tracking performance compared to the hand-tuned approach.
2502.04554
Unifying and Optimizing Data Values for Selection via Sequential-Decision-Making
cs.AI
Data selection has emerged as a crucial downstream application of data valuation. While existing data valuation methods have shown promise in selection tasks, the theoretical foundations and full potential of using data values for selection remain largely unexplored. In this work, we first demonstrate that data values applied for selection can be naturally reformulated as a sequential-decision-making problem, where the optimal data value can be derived through dynamic programming. We show this framework unifies and reinterprets existing methods like Data Shapley through the lens of approximate dynamic programming, specifically as myopic reward function approximations to this sequential problem. Furthermore, we analyze how sequential data selection optimality is affected when the ground-truth utility function exhibits monotonic submodularity with curvature. To address the computational challenges in obtaining optimal data values, we propose an efficient approximation scheme using learned bipartite graphs as surrogate utility models, ensuring greedy selection is still optimal when the surrogate utility is correctly specified and learned. Extensive experiments demonstrate the effectiveness of our approach across diverse datasets.
2502.04555
Decomposing Multivariate Information Rates in Networks of Random Processes
stat.ME cs.IT math.IT
The Partial Information Decomposition (PID) framework has emerged as a powerful tool for analyzing high-order interdependencies in complex network systems. However, its application to dynamic processes remains challenging due to the implicit assumption of memorylessness, which often falls in real-world scenarios. In this work, we introduce the framework of Partial Information Rate Decomposition (PIRD) that extends PID to random processes with temporal correlations. By leveraging mutual information rate (MIR) instead of mutual information (MI), our approach decomposes the dynamic information shared by multivariate random processes into unique, redundant, and synergistic contributions obtained aggregating information rate atoms in a principled manner. To solve PIRD, we define a pointwise redundancy rate function based on the minimum MI principle applied locally in the frequency-domain representation of the processes. The framework is validated in benchmark simulations of Gaussian systems, demonstrating its advantages over traditional PID in capturing temporal correlations and showing how the spectral representation may reveal scale-specific higher-order interactions that are obscured in the time domain. Furthermore, we apply PIRD to a physiological network comprising cerebrovascular and cardiovascular variables, revealing frequency-dependent redundant information exchange during a protocol of postural stress. Our results highlight the necessity of accounting for the full temporal statistical structure and spectral content of vector random processes to meaningfully perform information decomposition in network systems with dynamic behavior such as those typically encountered in neuroscience and physiology.
2502.04556
TruthFlow: Truthful LLM Generation via Representation Flow Correction
cs.CL cs.AI cs.LG
Large language models (LLMs) are known to struggle with consistently generating truthful responses. While various representation intervention techniques have been proposed, these methods typically apply a universal representation correction vector to all input queries, limiting their effectiveness against diverse queries in practice. In this study, we introduce TruthFlow, a novel method that leverages the Flow Matching technique for query-specific truthful representation correction. Specifically, TruthFlow first uses a flow model to learn query-specific correction vectors that transition representations from hallucinated to truthful states. Then, during inference, the trained flow model generates these correction vectors to enhance the truthfulness of LLM outputs. Experimental results demonstrate that TruthFlow significantly improves performance on open-ended generation tasks across various advanced LLMs evaluated on TruthfulQA. Moreover, the trained TruthFlow model exhibits strong transferability, performing effectively on other unseen hallucination benchmarks.
2502.04557
Speeding up Speculative Decoding via Approximate Verification
cs.LG cs.IT math.IT
Speculative Decoding (SD) is a recently proposed technique for faster inference using Large Language Models (LLMs). SD operates by using a smaller draft LLM for autoregressively generating a sequence of tokens and a larger target LLM for parallel verification to ensure statistical consistency. However, periodic parallel calls to the target LLM for verification prevent SD from achieving even lower latencies. We propose SPRINTER, which utilizes a low-complexity verifier trained to predict if tokens generated from a draft LLM would be accepted by the target LLM. By performing approximate sequential verification, SPRINTER does not require verification by the target LLM and is only invoked when a token is deemed unacceptable. This leads to reducing the number of calls to the larger LLM and can achieve further speedups. We present a theoretical analysis of SPRINTER, examining the statistical properties of the generated tokens, as well as the expected reduction in latency as a function of the verifier. We evaluate SPRINTER on several datasets and model pairs, demonstrating that approximate verification can still maintain high quality generation while further reducing latency. For instance, on Wiki-Summaries dataset, SPRINTER achieves a 1.7x latency speedup and requires 8.3x fewer flops relative to SD, while still generating high-quality responses when using GPT2-Small and GPT2-XL as draft/target models.
2502.04558
Probing a Vision-Language-Action Model for Symbolic States and Integration into a Cognitive Architecture
cs.RO cs.AI
Vision-language-action (VLA) models hold promise as generalist robotics solutions by translating visual and linguistic inputs into robot actions, yet they lack reliability due to their black-box nature and sensitivity to environmental changes. In contrast, cognitive architectures (CA) excel in symbolic reasoning and state monitoring but are constrained by rigid predefined execution. This work bridges these approaches by probing OpenVLA's hidden layers to uncover symbolic representations of object properties, relations, and action states, enabling integration with a CA for enhanced interpretability and robustness. Through experiments on LIBERO-spatial pick-and-place tasks, we analyze the encoding of symbolic states across different layers of OpenVLA's Llama backbone. Our probing results show consistently high accuracies (> 0.90) for both object and action states across most layers, though contrary to our hypotheses, we did not observe the expected pattern of object states being encoded earlier than action states. We demonstrate an integrated DIARC-OpenVLA system that leverages these symbolic representations for real-time state monitoring, laying the foundation for more interpretable and reliable robotic manipulation.
2502.04562
Mixture of neural operator experts for learning boundary conditions and model selection
cs.LG cs.NA math.NA physics.flu-dyn
While Fourier-based neural operators are best suited to learning mappings between functions on periodic domains, several works have introduced techniques for incorporating non trivial boundary conditions. However, all previously introduced methods have restrictions that limit their applicability. In this work, we introduce an alternative approach to imposing boundary conditions inspired by volume penalization from numerical methods and Mixture of Experts (MoE) from machine learning. By introducing competing experts, the approach additionally allows for model selection. To demonstrate the method, we combine a spatially conditioned MoE with the Fourier based, Modal Operator Regression for Physics (MOR-Physics) neural operator and recover a nonlinear operator on a disk and quarter disk. Next, we extract a large eddy simulation (LES) model from direct numerical simulation of channel flow and show the domain decomposition provided by our approach. Finally, we train our LES model with Bayesian variational inference and obtain posterior predictive samples of flow far past the DNS simulation time horizon.
2502.04563
WaferLLM: A Wafer-Scale LLM Inference System
cs.LG cs.AI cs.AR cs.DC cs.ET
Emerging AI accelerators increasingly adopt wafer-scale manufacturing technologies, integrating hundreds of thousands of AI cores in a mesh-based architecture with large distributed on-chip memory (tens of GB in total) and ultra-high on-chip memory bandwidth (tens of PB/s). However, current LLM inference systems, optimized for shared memory architectures like GPUs, fail to fully exploit these accelerators. We introduce WaferLLM, the first wafer-scale LLM inference system. WaferLLM is guided by a novel PLMR model (pronounced as "Plummer") that captures the unique hardware characteristics of wafer-scale architectures. Leveraging this model, WaferLLM pioneers wafer-scale LLM parallelism, optimizing the utilization of hundreds of thousands of on-chip cores. It also introduces MeshGEMM and MeshGEMV, the first GEMM and GEMV implementations designed to scale effectively on wafer-scale accelerators. Evaluations show that WaferLLM achieves 200$\times$ better wafer-scale accelerator utilization than state-of-the-art systems. On a commodity wafer-scale accelerator, WaferLLM delivers 606$\times$ faster and 22$\times$ more energy-efficient GEMV compared to an advanced GPU. For LLMs, based on 16-bit data type, WaferLLM achieves 2700 toks/sec/req decode speed on Llama3-8B model and 840 toks/sec/req decode speed on Qwen2-72B model, which enables 39$\times$ faster decoding with 1.7$\times$ better energy efficiency. We anticipate these numbers will grow significantly as wafer-scale AI models, software, and hardware continue to mature.
2502.04564
My LLM might Mimic AAE -- But When Should it?
cs.CL
We examine the representation of African American English (AAE) in large language models (LLMs), exploring (a) the perceptions Black Americans have of how effective these technologies are at producing authentic AAE, and (b) in what contexts Black Americans find this desirable. Through both a survey of Black Americans ($n=$ 104) and annotation of LLM-produced AAE by Black Americans ($n=$ 228), we find that Black Americans favor choice and autonomy in determining when AAE is appropriate in LLM output. They tend to prefer that LLMs default to communicating in Mainstream U.S. English in formal settings, with greater interest in AAE production in less formal settings. When LLMs were appropriately prompted and provided in context examples, our participants found their outputs to have a level of AAE authenticity on par with transcripts of Black American speech. Select code and data for our project can be found here: https://github.com/smelliecat/AAEMime.git
2502.04565
Private Federated Learning In Real World Application -- A Case Study
cs.LG cs.CR
This paper presents an implementation of machine learning model training using private federated learning (PFL) on edge devices. We introduce a novel framework that uses PFL to address the challenge of training a model using users' private data. The framework ensures that user data remain on individual devices, with only essential model updates transmitted to a central server for aggregation with privacy guarantees. We detail the architecture of our app selection model, which incorporates a neural network with attention mechanisms and ambiguity handling through uncertainty management. Experiments conducted through off-line simulations and on device training demonstrate the feasibility of our approach in real-world scenarios. Our results show the potential of PFL to improve the accuracy of an app selection model by adapting to changes in user behavior over time, while adhering to privacy standards. The insights gained from this study are important for industries looking to implement PFL, offering a robust strategy for training a predictive model directly on edge devices while ensuring user data privacy.
2502.04566
An Optimized YOLOv5 Based Approach For Real-time Vehicle Detection At Road Intersections Using Fisheye Cameras
cs.CV
Real time vehicle detection is a challenging task for urban traffic surveillance. Increase in urbanization leads to increase in accidents and traffic congestion in junction areas resulting in delayed travel time. In order to solve these problems, an intelligent system utilizing automatic detection and tracking system is significant. But this becomes a challenging task at road intersection areas which require a wide range of field view. For this reason, fish eye cameras are widely used in real time vehicle detection purpose to provide large area coverage and 360 degree view at junctions. However, it introduces challenges such as light glare from vehicles and street lights, shadow, non-linear distortion, scaling issues of vehicles and proper localization of small vehicles. To overcome each of these challenges, a modified YOLOv5 object detection scheme is proposed. YOLOv5 is a deep learning oriented convolutional neural network (CNN) based object detection method. The proposed scheme for detecting vehicles in fish-eye images consists of a light-weight day-night CNN classifier so that two different solutions can be implemented to address the day-night detection issues. Furthurmore, challenging instances are upsampled in the dataset for proper localization of vehicles and later on the detection model is ensembled and trained in different combination of vehicle datasets for better generalization, detection and accuracy. For testing, a real world fisheye dataset provided by the Video and Image Processing (VIP) Cup organizer ISSD has been used which includes images from video clips of different fisheye cameras at junction of different cities during day and night time. Experimental results show that our proposed model has outperformed the YOLOv5 model on the dataset by 13.7% mAP @ 0.5.
2502.04567
Preference Optimization via Contrastive Divergence: Your Reward Model is Secretly an NLL Estimator
cs.AI
Existing studies on preference optimization (PO) have centered on constructing pairwise preference data following simple heuristics, such as maximizing the margin between preferred and dispreferred completions based on human (or AI) ranked scores. However, none of these heuristics has a full theoretical justification. In this work, we develop a novel PO framework that provides theoretical guidance to effectively sample dispreferred completions. To achieve this, we formulate PO as minimizing the negative log-likelihood (NLL) of a probability model and propose to estimate its normalization constant via a sampling strategy. As we will demonstrate, these estimative samples can act as dispreferred completions in PO. We then select contrastive divergence (CD) as the sampling strategy, and propose a novel MC-PO algorithm that applies the Monte Carlo (MC) kernel from CD to sample hard negatives w.r.t. the parameterized reward model. Finally, we propose the OnMC-PO algorithm, an extension of MC-PO to the online setting. On popular alignment benchmarks, MC-PO outperforms existing SOTA baselines, and OnMC-PO leads to further improvement.
2502.04568
Learning Semantics-aware Search Operators for Genetic Programming
cs.LG cs.NE
Fitness landscapes in test-based program synthesis are known to be extremely rugged, with even minimal modifications of programs often leading to fundamental changes in their behavior and, consequently, fitness values. Relying on fitness as the only guidance in iterative search algorithms like genetic programming is thus unnecessarily limiting, especially when combined with purely syntactic search operators that are agnostic about their impact on program behavior. In this study, we propose a semantics-aware search operator that steers the search towards candidate programs that are valuable not only actually (high fitness) but also only potentially, i.e. are likely to be turned into high-quality solutions even if their current fitness is low. The key component of the method is a graph neural network that learns to model the interactions between program instructions and processed data, and produces a saliency map over graph nodes that represents possible search decisions. When applied to a suite of symbolic regression benchmarks, the proposed method outperforms conventional tree-based genetic programming and the ablated variant of the method.
2502.04573
Zero-shot Meta-learning for Tabular Prediction Tasks with Adversarially Pre-trained Transformer
cs.LG cs.AI
We present an Adversarially Pre-trained Transformer (APT) that is able to perform zero-shot meta-learning on tabular prediction tasks without pre-training on any real-world dataset, extending on the recent development of Prior-Data Fitted Networks (PFNs) and TabPFN. Specifically, APT is pre-trained with adversarial synthetic data agents, who continue to shift their underlying data generating distribution and deliberately challenge the model with different synthetic datasets. In addition, we propose a mixture block architecture that is able to handle classification tasks with arbitrary number of classes, addressing the class size limitation -- a crucial weakness of prior deep tabular zero-shot learners. In experiments, we show that our framework matches state-of-the-art performance on small classification tasks without filtering on dataset characteristics such as number of classes and number of missing values, while maintaining an average runtime under one second. On common benchmark dataset suites in both classification and regression, we show that adversarial pre-training was able to enhance TabPFN's performance. In our analysis, we demonstrate that the adversarial synthetic data agents were able to generate a more diverse collection of data compared to the ordinary random generator in TabPFN. In addition, we demonstrate that our mixture block neural design has improved generalizability and greatly accelerated pre-training.
2502.04574
Dark Brain Energy: Toward an Integrative Model of Spontaneous Slow Oscillations
q-bio.NC cs.IT math.IT stat.AP
Neural oscillations facilitate the functioning of the human brain in spatial and temporal dimensions at various frequencies. These oscillations feature a universal frequency architecture that is governed by brain anatomy, ensuring frequency specificity remains invariant across different measurement techniques. Initial magnetic resonance imaging (MRI) methodology constrained functional MRI (fMRI) investigations to a singular frequency range, thereby neglecting the frequency characteristics inherent in blood oxygen level-dependent oscillations. With advancements in MRI technology, it has become feasible to decode intricate brain activities via multi-band frequency analysis (MBFA). During the past decade, the utilization of MBFA in fMRI studies has surged, unveiling frequency-dependent characteristics of spontaneous slow oscillations (SSOs) believed to base dark energy in the brain. There remains a dearth of conclusive insights and hypotheses pertaining to the properties and functionalities of SSOs in distinct bands. We surveyed the SSO MBFA studies during the past 15 years to delineate the attributes of SSOs and enlighten their correlated functions. We further proposed a model to elucidate the hierarchical organization of multi-band SSOs by integrating their function, aimed at bridging theoretical gaps and guiding future MBFA research endeavors.
2502.04575
Complexity Analysis of Normalizing Constant Estimation: from Jarzynski Equality to Annealed Importance Sampling and beyond
stat.ML cs.LG cs.NA math.NA physics.comp-ph stat.CO
Given an unnormalized probability density $\pi\propto\mathrm{e}^{-V}$, estimating its normalizing constant $Z=\int_{\mathbb{R}^d}\mathrm{e}^{-V(x)}\mathrm{d}x$ or free energy $F=-\log Z$ is a crucial problem in Bayesian statistics, statistical mechanics, and machine learning. It is challenging especially in high dimensions or when $\pi$ is multimodal. To mitigate the high variance of conventional importance sampling estimators, annealing-based methods such as Jarzynski equality and annealed importance sampling are commonly adopted, yet their quantitative complexity guarantees remain largely unexplored. We take a first step toward a non-asymptotic analysis of annealed importance sampling. In particular, we derive an oracle complexity of $\widetilde{O}\left(\frac{d\beta^2{\mathcal{A}}^2}{\varepsilon^4}\right)$ for estimating $Z$ within $\varepsilon$ relative error with high probability, where $\beta$ is the smoothness of $V$ and $\mathcal{A}$ denotes the action of a curve of probability measures interpolating $\pi$ and a tractable reference distribution. Our analysis, leveraging Girsanov theorem and optimal transport, does not explicitly require isoperimetric assumptions on the target distribution. Finally, to tackle the large action of the widely used geometric interpolation of probability distributions, we propose a new normalizing constant estimation algorithm based on reverse diffusion samplers and establish a framework for analyzing its complexity.
2502.04576
Self-Regulation and Requesting Interventions
cs.LG cs.CL
Human intelligence involves metacognitive abilities like self-regulation, recognizing limitations, and seeking assistance only when needed. While LLM Agents excel in many domains, they often lack this awareness. Overconfident agents risk catastrophic failures, while those that seek help excessively hinder efficiency. A key challenge is enabling agents with a limited intervention budget $C$ is to decide when to request assistance. In this paper, we propose an offline framework that trains a "helper" policy to request interventions, such as more powerful models or test-time compute, by combining LLM-based process reward models (PRMs) with tabular reinforcement learning. Using state transitions collected offline, we score optimal intervention timing with PRMs and train the helper model on these labeled trajectories. This offline approach significantly reduces costly intervention calls during training. Furthermore, the integration of PRMs with tabular RL enhances robustness to off-policy data while avoiding the inefficiencies of deep RL. We empirically find that our method delivers optimal helper behavior.
2502.04577
Position-aware Automatic Circuit Discovery
cs.LG cs.CL
A widely used strategy to discover and understand language model mechanisms is circuit analysis. A circuit is a minimal subgraph of a model's computation graph that executes a specific task. We identify a gap in existing circuit discovery methods: they assume circuits are position-invariant, treating model components as equally relevant across input positions. This limits their ability to capture cross-positional interactions or mechanisms that vary across positions. To address this gap, we propose two improvements to incorporate positionality into circuits, even on tasks containing variable-length examples. First, we extend edge attribution patching, a gradient-based method for circuit discovery, to differentiate between token positions. Second, we introduce the concept of a dataset schema, which defines token spans with similar semantics across examples, enabling position-aware circuit discovery in datasets with variable length examples. We additionally develop an automated pipeline for schema generation and application using large language models. Our approach enables fully automated discovery of position-sensitive circuits, yielding better trade-offs between circuit size and faithfulness compared to prior work.
2502.04580
Technical Debt in In-Context Learning: Diminishing Efficiency in Long Context
cs.LG cs.AI
Transformers have demonstrated remarkable in-context learning (ICL) capabilities, adapting to new tasks by simply conditioning on demonstrations without parameter updates. Compelling empirical and theoretical evidence suggests that ICL, as a general-purpose learner, could outperform task-specific models. However, it remains unclear to what extent the transformers optimally learn in-context compared to principled learning algorithms. To bridge this gap, we introduce a new framework for quantifying optimality of ICL as a learning algorithm in stylized settings. Our findings reveal a striking dichotomy: while ICL initially matches the efficiency of a Bayes optimal estimator, its efficiency significantly deteriorates in long context. Through an information-theoretic analysis, we show that the diminishing efficiency is inherent to ICL. These results clarify the trade-offs in adopting ICL as a universal problem solver, motivating a new generation of on-the-fly adaptive methods without the diminishing efficiency.
2502.04582
The Mini Wheelbot: A Testbed for Learning-based Balancing, Flips, and Articulated Driving
cs.RO cs.SY eess.SY math.OC
The Mini Wheelbot is a balancing, reaction wheel unicycle robot designed as a testbed for learning-based control. It is an unstable system with highly nonlinear yaw dynamics, non-holonomic driving, and discrete contact switches in a small, powerful, and rugged form factor. The Mini Wheelbot can use its wheels to stand up from any initial orientation - enabling automatic environment resets in repetitive experiments and even challenging half flips. We illustrate the effectiveness of the Mini Wheelbot as a testbed by implementing two popular learning-based control algorithms. First, we showcase Bayesian optimization for tuning the balancing controller. Second, we use imitation learning from an expert nonlinear MPC that uses gyroscopic effects to reorient the robot and can track higher-level velocity and orientation commands. The latter allows the robot to drive around based on user commands - for the first time in this class of robots. The Mini Wheelbot is not only compelling for testing learning-based control algorithms, but it is also just fun to work with, as demonstrated in the video of our experiments.
2502.04583
Overcoming Fake Solutions in Semi-Dual Neural Optimal Transport: A Smoothing Approach for Learning the Optimal Transport Plan
cs.LG
We address the convergence problem in learning the Optimal Transport (OT) map, where the OT Map refers to a map from one distribution to another while minimizing the transport cost. Semi-dual Neural OT, a widely used approach for learning OT Maps with neural networks, often generates fake solutions that fail to transfer one distribution to another accurately. We identify a sufficient condition under which the max-min solution of Semi-dual Neural OT recovers the true OT Map. Moreover, to address cases when this sufficient condition is not satisfied, we propose a novel method, OTP, which learns both the OT Map and the Optimal Transport Plan, representing the optimal coupling between two distributions. Under sharp assumptions on the distributions, we prove that our model eliminates the fake solution issue and correctly solves the OT problem. Our experiments show that the OTP model recovers the optimal transport map where existing methods fail and outperforms current OT-based models in image-to-image translation tasks. Notably, the OTP model can learn stochastic transport maps when deterministic OT Maps do not exist, such as one-to-many tasks like colorization.
2502.04584
Joint State and Noise Covariance Estimation
cs.RO math.OC
This paper tackles the problem of jointly estimating the noise covariance matrix alongside primary parameters (such as poses and points) from measurements corrupted by Gaussian noise. In such settings, the noise covariance matrix determines the weights assigned to individual measurements in the least squares problem. We show that the joint problem exhibits a convex structure and provide a full characterization of the optimal noise covariance estimate (with analytical solutions) within joint maximum a posteriori and likelihood frameworks and several variants. Leveraging this theoretical result, we propose two novel algorithms that jointly estimate the primary parameters and the noise covariance matrix. To validate our approach, we conduct extensive experiments across diverse scenarios and offer practical insights into their application in robotics and computer vision estimation problems with a particular focus on SLAM.
2502.04586
Automatic Ply Partitioning for Laminar Composite Process Planning
math.OC cs.CE
This work introduces an automated ply partitioning strategy for large-scale laminar composite manufacturing. It specifically targets the problem of fabricating large plies from available spooled materials, while minimizing the adverse effects on part quality. The proposed method inserts fiber-aligned seams sequentially until each resulting sub-ply can be manufactured from available materials, while simultaneously enforcing constraints to avoid quality issues induced by the stacking of seams across multiple plies. Leveraging the developable nature of individual plies, the partitioning problem is cast as a sequence of one-dimensional piecewise linear optimization problems, thus allowing for efficient local optimization via linear programming. We experimentally demonstrate that coupling the local search with a greedy global search produces the same results as an exhaustive search. The resulting automated method provides an efficient and robust alternative to the existing trial-and-error approach, and can be readily integrated into state-of-the-art composite design workflows. In addition, this formulation enables the inclusion of common constraints regarding laminate thickness tolerance, sub-ply geometry, stay-out zones, material wastage, etc. The efficacy of the proposed method is demonstrated through its application to the surface of an airplane wing and to the body panels of an armored vehicle, each subject to various performance and manufacturing-related geometric constraints.
2502.04591
Rethinking Oversmoothing in Graph Neural Networks: A Rank-Based Perspective
cs.LG cs.AI stat.ML
Oversmoothing is a fundamental challenge in graph neural networks (GNNs): as the number of layers increases, node embeddings become increasingly similar, and model performance drops sharply. Traditionally, oversmoothing has been quantified using metrics that measure the similarity of neighbouring node features, such as the Dirichlet energy. While these metrics are related to oversmoothing, we argue they have critical limitations and fail to reliably capture oversmoothing in realistic scenarios. For instance, they provide meaningful insights only for very deep networks and under somewhat strict conditions on the norm of network weights and feature representations. As an alternative, we propose measuring oversmoothing by examining the numerical or effective rank of the feature representations. We provide theoretical support for this approach, demonstrating that the numerical rank of feature representations converges to one for a broad family of nonlinear activation functions under the assumption of nonnegative trained weights. To the best of our knowledge, this is the first result that proves the occurrence of oversmoothing without assumptions on the boundedness of the weight matrices. Along with the theoretical findings, we provide extensive numerical evaluation across diverse graph architectures. Our results show that rank-based metrics consistently capture oversmoothing, whereas energy-based metrics often fail. Notably, we reveal that a significant drop in the rank aligns closely with performance degradation, even in scenarios where energy metrics remain unchanged.
2502.04592
CAMEF: Causal-Augmented Multi-Modality Event-Driven Financial Forecasting by Integrating Time Series Patterns and Salient Macroeconomic Announcements
cs.LG cs.AI cs.CE
Accurately forecasting the impact of macroeconomic events is critical for investors and policymakers. Salient events like monetary policy decisions and employment reports often trigger market movements by shaping expectations of economic growth and risk, thereby establishing causal relationships between events and market behavior. Existing forecasting methods typically focus either on textual analysis or time-series modeling, but fail to capture the multi-modal nature of financial markets and the causal relationship between events and price movements. To address these gaps, we propose CAMEF (Causal-Augmented Multi-Modality Event-Driven Financial Forecasting), a multi-modality framework that effectively integrates textual and time-series data with a causal learning mechanism and an LLM-based counterfactual event augmentation technique for causal-enhanced financial forecasting. Our contributions include: (1) a multi-modal framework that captures causal relationships between policy texts and historical price data; (2) a new financial dataset with six types of macroeconomic releases from 2008 to April 2024, and high-frequency real trading data for five key U.S. financial assets; and (3) an LLM-based counterfactual event augmentation strategy. We compare CAMEF to state-of-the-art transformer-based time-series and multi-modal baselines, and perform ablation studies to validate the effectiveness of the causal learning mechanism and event types.
2502.04593
The $\alpha$-Alternator: Dynamic Adaptation To Varying Noise Levels In Sequences Using The Vendi Score For Improved Robustness and Performance
cs.LG cs.AI cs.NE stat.ML
Current state-of-the-art dynamical models, such as Mamba, assume the same level of noisiness for all elements of a given sequence, which limits their performance on noisy temporal data. In this paper, we introduce the $\alpha$-Alternator, a novel generative model for time-dependent data that dynamically adapts to the complexity introduced by varying noise levels in sequences. The $\alpha$-Alternator leverages the Vendi Score (VS), a flexible similarity-based diversity metric, to adjust, at each time step $t$, the influence of the sequence element at time $t$ and the latent representation of the dynamics up to that time step on the predicted future dynamics. This influence is captured by a parameter that is learned and shared across all sequences in a given dataset. The sign of this parameter determines the direction of influence. A negative value indicates a noisy dataset, where a sequence element that increases the VS is considered noisy, and the model relies more on the latent history when processing that element. Conversely, when the parameter is positive, a sequence element that increases the VS is considered informative, and the $\alpha$-Alternator relies more on this new input than on the latent history when updating its predicted latent dynamics. The $\alpha$-Alternator is trained using a combination of observation masking and Alternator loss minimization. Masking simulates varying noise levels in sequences, enabling the model to be more robust to these fluctuations and improving its performance in trajectory prediction, imputation, and forecasting. Our experimental results demonstrate that the $\alpha$-Alternator outperforms both Alternators and state-of-the-art state-space models across neural decoding and time-series forecasting benchmarks.
2502.04595
A Fractional-Order Nonlinear Backstepping Controller Design for Current-Controlled Maglev System
eess.SY cs.SY
The magnetic levitation system (Maglev) is a nonlinear system by which an object is suspended with no support other than magnetic fields. The main control perspective of the Maglev system is to levitate a steel ball in air by the electromagnetic force. However, the Maglev system has highly nonlinear dynamics which is inconvenient in the sense of sensitive control/regulation of its nonlinear dynamics. In this paper, the nonlinear backstepping controller based on the fractional-order derivative is proposed for the control of the nonlinear current-controlled Maglev system. After, the system dynamics and fractional-order backstepping controller design are given, the asymptotic stability of the closed-loop system is proved by employing the Lyapunov theory. Some computer-based numerical experiments are carried out to show the effectiveness of the proposed controller for the control of Maglev system.
2502.04597
Multiscale style transfer based on a Laplacian pyramid for traditional Chinese painting
cs.CV
Style transfer is adopted to synthesize appealing stylized images that preserve the structure of a content image but carry the pattern of a style image. Many recently proposed style transfer methods use only western oil paintings as style images to achieve image stylization. As a result, unnatural messy artistic effects are produced in stylized images when using these methods to directly transfer the patterns of traditional Chinese paintings, which are composed of plain colors and abstract objects. Moreover, most of them work only at the original image scale and thus ignore multiscale image information during training. In this paper, we present a novel effective multiscale style transfer method based on Laplacian pyramid decomposition and reconstruction, which can transfer unique patterns of Chinese paintings by learning different image features at different scales. In the first stage, the holistic patterns are transferred at low resolution by adopting a Style Transfer Base Network. Then, the details of the content and style are gradually enhanced at higher resolutions by a Detail Enhancement Network with an edge information selection (EIS) module in the second stage. The effectiveness of our method is demonstrated through the generation of appealing high-quality stylization results and a comparison with some state-of-the-art style transfer methods. Datasets and codes are available at https://github.com/toby-katakuri/LP_StyleTransferNet.
2502.04600
Cooperative Payload Estimation by a Team of Mocobots
cs.RO
Consider the following scenario: a human guides multiple mobile manipulators to grasp a common payload. For subsequent high-performance autonomous manipulation of the payload by the mobile manipulator team, or for collaborative manipulation with the human, the robots should be able to discover where the other robots are attached to the payload, as well as the payload's mass and inertial properties. In this paper, we describe a method for the robots to autonomously discover this information. The robots cooperatively manipulate the payload, and the twist, twist derivative, and wrench data at their grasp frames are used to estimate the transformation matrices between the grasp frames, the location of the payload's center of mass, and the payload's inertia matrix. The method is validated experimentally with a team of three mobile cobots, or mocobots.
2502.04601
LATTEO: A Framework to Support Learning Asynchronously Tempered with Trusted Execution and Obfuscation
cs.CR cs.LG
The privacy vulnerabilities of the federated learning (FL) paradigm, primarily caused by gradient leakage, have prompted the development of various defensive measures. Nonetheless, these solutions have predominantly been crafted for and assessed in the context of synchronous FL systems, with minimal focus on asynchronous FL. This gap arises in part due to the unique challenges posed by the asynchronous setting, such as the lack of coordinated updates, increased variability in client participation, and the potential for more severe privacy risks. These concerns have stymied the adoption of asynchronous FL. In this work, we first demonstrate the privacy vulnerabilities of asynchronous FL through a novel data reconstruction attack that exploits gradient updates to recover sensitive client data. To address these vulnerabilities, we propose a privacy-preserving framework that combines a gradient obfuscation mechanism with Trusted Execution Environments (TEEs) for secure asynchronous FL aggregation at the network edge. To overcome the limitations of conventional enclave attestation, we introduce a novel data-centric attestation mechanism based on Multi-Authority Attribute-Based Encryption. This mechanism enables clients to implicitly verify TEE-based aggregation services, effectively handle on-demand client participation, and scale seamlessly with an increasing number of asynchronous connections. Our gradient obfuscation mechanism reduces the structural similarity index of data reconstruction by 85% and increases reconstruction error by 400%, while our framework improves attestation efficiency by lowering average latency by up to 1500% compared to RA-TLS, without additional overhead.
2502.04602
Extracting and Understanding the Superficial Knowledge in Alignment
cs.CL cs.AI
Alignment of large language models (LLMs) with human values and preferences, often achieved through fine-tuning based on human feedback, is essential for ensuring safe and responsible AI behaviors. However, the process typically requires substantial data and computation resources. Recent studies have revealed that alignment might be attainable at lower costs through simpler methods, such as in-context learning. This leads to the question: Is alignment predominantly superficial? In this paper, we delve into this question and provide a quantitative analysis. We formalize the concept of superficial knowledge, defining it as knowledge that can be acquired through easily token restyling, without affecting the model's ability to capture underlying causal relationships between tokens. We propose a method to extract and isolate superficial knowledge from aligned models, focusing on the shallow modifications to the final token selection process. By comparing models augmented only with superficial knowledge to fully aligned models, we quantify the superficial portion of alignment. Our findings reveal that while superficial knowledge constitutes a significant portion of alignment, particularly in safety and detoxification tasks, it is not the whole story. Tasks requiring reasoning and contextual understanding still rely on deeper knowledge. Additionally, we demonstrate two practical advantages of isolated superficial knowledge: (1) it can be transferred between models, enabling efficient offsite alignment of larger models using extracted superficial knowledge from smaller models, and (2) it is recoverable, allowing for the restoration of alignment in compromised models without sacrificing performance.
2502.04609
Force interaction, modeling and soft tissue deformation during reciprocating insertion of multi-part probe
cs.RO cs.SY eess.SY
The bio-inspired engineering of ovipositing wasps, which employ a reciprocating motion for soft tissue insertion, offers potential advantages in reducing insertion force and minimizing tissue damage. However, the underlying mechanisms of tissue interaction and sparing are not fully understood. In this study, we aim to investigate a multi-part probe designed to mimic the reciprocating motion of ovipositors. A reciprocal insertion model was developed to study the interaction between the probe and soft tissue, and experimental testing was conducted using a force sensor and laser optical technique to gain insights into interacting forces and tissue deformation. The results reveal that during the cutting phase of reciprocal motion, the peak force and average displacement of the soft substrate were approximately 19% and 20% lower, respectively, compared to direct insertion at an overall probe velocity of 1 mm/s. This study presents a novel approach combining mechanical modeling and experimental analysis to explore the force mechanics of the reciprocating insertion method, providing a better understanding of the interaction between the probe and soft tissue.
2502.04615
Neural Clustering for Prefractured Mesh Generation in Real-time Object Destruction
cs.CV cs.GR
Prefracture method is a practical implementation for real-time object destruction that is hardly achievable within performance constraints, but can produce unrealistic results due to its heuristic nature. To mitigate it, we approach the clustering of prefractured mesh generation as an unordered segmentation on point cloud data, and propose leveraging the deep neural network trained on a physics-based dataset. Our novel paradigm successfully predicts the structural weakness of object that have been limited, exhibiting ready-to-use results with remarkable quality.
2502.04623
HetSSNet: Spatial-Spectral Heterogeneous Graph Learning Network for Panchromatic and Multispectral Images Fusion
cs.CV
Remote sensing pansharpening aims to reconstruct spatial-spectral properties during the fusion of panchromatic (PAN) images and low-resolution multi-spectral (LR-MS) images, finally generating the high-resolution multi-spectral (HR-MS) images. In the mainstream modeling strategies, i.e., CNN and Transformer, the input images are treated as the equal-sized grid of pixels in the Euclidean space. They have limitations in facing remote sensing images with irregular ground objects. Graph is the more flexible structure, however, there are two major challenges when modeling spatial-spectral properties with graph: \emph{1) constructing the customized graph structure for spatial-spectral relationship priors}; \emph{2) learning the unified spatial-spectral representation through the graph}. To address these challenges, we propose the spatial-spectral heterogeneous graph learning network, named \textbf{HetSSNet}. Specifically, HetSSNet initially constructs the heterogeneous graph structure for pansharpening, which explicitly describes pansharpening-specific relationships. Subsequently, the basic relationship pattern generation module is designed to extract the multiple relationship patterns from the heterogeneous graph. Finally, relationship pattern aggregation module is exploited to collaboratively learn unified spatial-spectral representation across different relationships among nodes with adaptive importance learning from local and global perspectives. Extensive experiments demonstrate the significant superiority and generalization of HetSSNet.
2502.04625
Phonetic Reconstruction of the Consonant System of Middle Chinese via Mixed Integer Optimization
cs.CL
This paper is concerned with phonetic reconstruction of the consonant system of Middle Chinese. We propose to cast the problem as a Mixed Integer Programming problem, which is able to automatically explore homophonic information from ancient rhyme dictionaries and phonetic information from modern Chinese dialects, the descendants of Middle Chinese. Numerical evaluation on a wide range of synthetic and real data demonstrates the effectiveness and robustness of the new method. We apply the method to information from Guangyun and 20 modern Chinese dialects to obtain a new phonetic reconstruction result. A linguistically-motivated discussion of this result is also provided.
2502.04628
AIQViT: Architecture-Informed Post-Training Quantization for Vision Transformers
cs.CV
Post-training quantization (PTQ) has emerged as a promising solution for reducing the storage and computational cost of vision transformers (ViTs). Recent advances primarily target at crafting quantizers to deal with peculiar activations characterized by ViTs. However, most existing methods underestimate the information loss incurred by weight quantization, resulting in significant performance deterioration, particularly in low-bit cases. Furthermore, a common practice in quantizing post-Softmax activations of ViTs is to employ logarithmic transformations, which unfortunately prioritize less informative values around zero. This approach introduces additional redundancies, ultimately leading to suboptimal quantization efficacy. To handle these, this paper proposes an innovative PTQ method tailored for ViTs, termed AIQViT (Architecture-Informed Post-training Quantization for ViTs). First, we design an architecture-informed low rank compensation mechanism, wherein learnable low-rank weights are introduced to compensate for the degradation caused by weight quantization. Second, we design a dynamic focusing quantizer to accommodate the unbalanced distribution of post-Softmax activations, which dynamically selects the most valuable interval for higher quantization resolution. Extensive experiments on five vision tasks, including image classification, object detection, instance segmentation, point cloud classification, and point cloud part segmentation, demonstrate the superiority of AIQViT over state-of-the-art PTQ methods.
2502.04630
High-Speed Dynamic 3D Imaging with Sensor Fusion Splatting
cs.CV cs.GR
Capturing and reconstructing high-speed dynamic 3D scenes has numerous applications in computer graphics, vision, and interdisciplinary fields such as robotics, aerodynamics, and evolutionary biology. However, achieving this using a single imaging modality remains challenging. For instance, traditional RGB cameras suffer from low frame rates, limited exposure times, and narrow baselines. To address this, we propose a novel sensor fusion approach using Gaussian splatting, which combines RGB, depth, and event cameras to capture and reconstruct deforming scenes at high speeds. The key insight of our method lies in leveraging the complementary strengths of these imaging modalities: RGB cameras capture detailed color information, event cameras record rapid scene changes with microsecond resolution, and depth cameras provide 3D scene geometry. To unify the underlying scene representation across these modalities, we represent the scene using deformable 3D Gaussians. To handle rapid scene movements, we jointly optimize the 3D Gaussian parameters and their temporal deformation fields by integrating data from all three sensor modalities. This fusion enables efficient, high-quality imaging of fast and complex scenes, even under challenging conditions such as low light, narrow baselines, or rapid motion. Experiments on synthetic and real datasets captured with our prototype sensor fusion setup demonstrate that our method significantly outperforms state-of-the-art techniques, achieving noticeable improvements in both rendering fidelity and structural accuracy.
2502.04632
Tight Bounds for Noisy Computation of High-Influence Functions, Connectivity, and Threshold
cs.DS cs.CC cs.IT math.IT
In the noisy query model, the (binary) return value of every query (possibly repeated) is independently flipped with some fixed probability $p \in (0, 1/2)$. In this paper, we obtain tight bounds on the noisy query complexity of several fundamental problems. Our first contribution is to show that any Boolean function with total influence $\Omega(n)$ has noisy query complexity $\Theta(n\log n)$. Previous works often focus on specific problems, and it is of great interest to have a characterization of noisy query complexity for general functions. Our result is the first noisy query complexity lower bound of this generality, beyond what was known for random Boolean functions [Reischuk and Schmeltz, FOCS 1991]. Our second contribution is to prove that Graph Connectivity has noisy query complexity $\Theta(n^2 \log n)$. In this problem, the goal is to determine whether an undirected graph is connected using noisy edge queries. While the upper bound can be achieved by a simple algorithm, no non-trivial lower bounds were known prior to this work. Last but not least, we determine the exact number of noisy queries (up to lower order terms) needed to solve the $k$-Threshold problem and the Counting problem. The $k$-Threshold problem asks to decide whether there are at least $k$ ones among $n$ bits, given noisy query access to the bits. We prove that $(1\pm o(1)) \frac{n\log (\min\{k,n-k+1\}/\delta)}{(1-2p)\log \frac{1-p}p}$ queries are both sufficient and necessary to achieve error probability $\delta = o(1)$. Previously, such a result was only known when $\min\{k,n-k+1\}=o(n)$ [Wang, Ghaddar, Zhu and Wang, arXiv 2024]. We also show a similar $(1\pm o(1)) \frac{n\log (\min\{k+1,n-k+1\}/\delta)}{(1-2p)\log \frac{1-p}p}$ bound for the Counting problem, where one needs to count the number of ones among $n$ bits given noisy query access and $k$ denotes the answer.
2502.04635
Exercise Specialists Evaluation of Robot-led Physical Therapy for People with Parkinsons Disease
cs.RO
Robot-led physical therapy (PT) offers a promising avenue to enhance the care provided by clinical exercise specialists (ES) and physical and occupational therapists to improve patients' adherence to prescribed exercises outside of a clinic, such as at home. Collaborative efforts among roboticists, ES, physical and occupational therapists, and patients are essential for developing interactive, personalized exercise systems that meet each stakeholder's needs. We conducted a user study in which 11 ES evaluated a novel robot-led PT system for people with Parkinson's disease (PD), introduced in [1], focusing on the system's perceived efficacy and acceptance. Utilizing a mixed-methods approach, including technology acceptance questionnaires, task load questionnaires, and semi-structured interviews, we gathered comprehensive insights into ES perspectives and experiences after interacting with the system. Findings reveal a broadly positive reception, which highlights the system's capacity to augment traditional PT for PD, enhance patient engagement, and ensure consistent exercise support. We also identified two key areas for improvement: incorporating more human-like feedback systems and increasing the robot's ease of use. This research emphasizes the value of incorporating robotic aids into PT for PD, offering insights that can guide the development of more effective and user-friendly rehabilitation technologies.
2502.04636
An Empirical Study of Code Obfuscation Practices in the Google Play Store
cs.CR cs.AI cs.SE
The Android ecosystem is vulnerable to issues such as app repackaging, counterfeiting, and piracy, threatening both developers and users. To mitigate these risks, developers often employ code obfuscation techniques. However, while effective in protecting legitimate applications, obfuscation also hinders security investigations as it is often exploited for malicious purposes. As such, it is important to understand code obfuscation practices in Android apps. In this paper, we analyze over 500,000 Android APKs from Google Play, spanning an eight-year period, to investigate the evolution and prevalence of code obfuscation techniques. First, we propose a set of classifiers to detect obfuscated code, tools, and techniques and then conduct a longitudinal analysis to identify trends. Our results show a 13% increase in obfuscation from 2016 to 2023, with ProGuard and Allatori as the most commonly used tools. We also show that obfuscation is more prevalent in top-ranked apps and gaming genres such as Casino apps. To our knowledge, this is the first large-scale study of obfuscation adoption in the Google Play Store, providing insights for developers and security analysts.
2502.04638
Learning Street View Representations with Spatiotemporal Contrast
cs.CV cs.AI
Street view imagery is extensively utilized in representation learning for urban visual environments, supporting various sustainable development tasks such as environmental perception and socio-economic assessment. However, it is challenging for existing image representations to specifically encode the dynamic urban environment (such as pedestrians, vehicles, and vegetation), the built environment (including buildings, roads, and urban infrastructure), and the environmental ambiance (such as the cultural and socioeconomic atmosphere) depicted in street view imagery to address downstream tasks related to the city. In this work, we propose an innovative self-supervised learning framework that leverages temporal and spatial attributes of street view imagery to learn image representations of the dynamic urban environment for diverse downstream tasks. By employing street view images captured at the same location over time and spatially nearby views at the same time, we construct contrastive learning tasks designed to learn the temporal-invariant characteristics of the built environment and the spatial-invariant neighborhood ambiance. Our approach significantly outperforms traditional supervised and unsupervised methods in tasks such as visual place recognition, socioeconomic estimation, and human-environment perception. Moreover, we demonstrate the varying behaviors of image representations learned through different contrastive learning objectives across various downstream tasks. This study systematically discusses representation learning strategies for urban studies based on street view images, providing a benchmark that enhances the applicability of visual data in urban science. The code is available at https://github.com/yonglleee/UrbanSTCL.
2502.04640
Building Rome with Convex Optimization
cs.RO cs.CV math.OC
Global bundle adjustment is made easy by depth prediction and convex optimization. We (i) propose a scaled bundle adjustment (SBA) formulation that lifts 2D keypoint measurements to 3D with learned depth, (ii) design an empirically tight convex semidfinite program (SDP) relaxation that solves SBA to certfiable global optimality, (iii) solve the SDP relaxations at extreme scale with Burer-Monteiro factorization and a CUDA-based trust-region Riemannian optimizer (dubbed XM), (iv) build a structure from motion (SfM) pipeline with XM as the optimization engine and show that XM-SfM dominates or compares favorably with existing SfM pipelines in terms of reconstruction quality while being faster, more scalable, and initialization-free.
2502.04642
Dynamic Incentive Selection for Hierarchical Convex Model Predictive Control
eess.SY cs.SY math.OC
In this paper, we discuss incentive design for hierarchical model predictive control (MPC) systems viewed as Stackelberg games. We consider a hierarchical MPC formulation where, given a lower-level convex MPC (LoMPC), the upper-level system solves a bilevel MPC (BiMPC) subject to the constraint that the lower-level system inputs are optimal for the LoMPC. Such hierarchical problems are challenging due to optimality constraints in the BiMPC formulation. We analyze an incentive Stackelberg game variation of the problem, where the BiMPC provides additional incentives for the LoMPC cost function, which grants the BiMPC influence over the LoMPC inputs. We show that for such problems, the BiMPC can be reformulated as a simpler optimization problem, and the optimal incentives can be iteratively computed without knowing the LoMPC cost function. We extend our formulation for the case of multiple LoMPCs and propose an algorithm that finds bounded suboptimal solutions for the BiMPC. We demonstrate our algorithm for a dynamic price control example, where an independent system operator (ISO) sets the electricity prices for electric vehicle (EV) charging with the goal of minimizing a social cost and satisfying electricity generation constraints. Notably, our method scales well to large EV population sizes.
2502.04643
Confidence Elicitation: A New Attack Vector for Large Language Models
cs.LG cs.CL cs.CR
A fundamental issue in deep learning has been adversarial robustness. As these systems have scaled, such issues have persisted. Currently, large language models (LLMs) with billions of parameters suffer from adversarial attacks just like their earlier, smaller counterparts. However, the threat models have changed. Previously, having gray-box access, where input embeddings or output logits/probabilities were visible to the user, might have been reasonable. However, with the introduction of closed-source models, no information about the model is available apart from the generated output. This means that current black-box attacks can only utilize the final prediction to detect if an attack is successful. In this work, we investigate and demonstrate the potential of attack guidance, akin to using output probabilities, while having only black-box access in a classification setting. This is achieved through the ability to elicit confidence from the model. We empirically show that the elicited confidence is calibrated and not hallucinated for current LLMs. By minimizing the elicited confidence, we can therefore increase the likelihood of misclassification. Our new proposed paradigm demonstrates promising state-of-the-art results on three datasets across two models (LLaMA-3-8B-Instruct and Mistral-7B-Instruct-V0.3) when comparing our technique to existing hard-label black-box attack methods that introduce word-level substitutions.
2502.04644
Agentic Reasoning: Reasoning LLMs with Tools for the Deep Research
cs.AI cs.CL
We introduce Agentic Reasoning, a framework that enhances large language model (LLM) reasoning by integrating external tool-using agents. Unlike conventional LLM-based reasoning approaches, which rely solely on internal inference, Agentic Reasoning dynamically engages web search, code execution, and structured reasoning-context memory to solve complex problems requiring deep research and multi-step logical deduction. Our framework introduces the Mind Map agent, which constructs a structured knowledge graph to track logical relationships, improving deductive reasoning. Additionally, the integration of web-search and coding agents enables real-time retrieval and computational analysis, enhancing reasoning accuracy and decision-making. Evaluations on PhD-level scientific reasoning (GPQA) and domain-specific deep research tasks demonstrate that our approach significantly outperforms existing models, including leading retrieval-augmented generation (RAG) systems and closed-source LLMs. Moreover, our results indicate that agentic reasoning improves expert-level knowledge synthesis, test-time scalability, and structured problem-solving. The code is at: https://github.com/theworldofagents/Agentic-Reasoning.
2502.04645
Cross-Encoder Rediscovers a Semantic Variant of BM25
cs.IR cs.AI
Neural Ranking Models (NRMs) have rapidly advanced state-of-the-art performance on information retrieval tasks. In this work, we investigate a Cross-Encoder variant of MiniLM to determine which relevance features it computes and where they are stored. We find that it employs a semantic variant of the traditional BM25 in an interpretable manner, featuring localized components: (1) Transformer attention heads that compute soft term frequency while controlling for term saturation and document length effects, and (2) a low-rank component of its embedding matrix that encodes inverse document frequency information for the vocabulary. This suggests that the Cross-Encoder uses the same fundamental mechanisms as BM25, but further leverages their capacity to capture semantics for improved retrieval performance. The granular understanding lays the groundwork for model editing to enhance model transparency, addressing safety concerns, and improving scalability in training and real-world applications.
2502.04646
Importance Sampling via Score-based Generative Models
cs.LG cs.AI
Importance sampling, which involves sampling from a probability density function (PDF) proportional to the product of an importance weight function and a base PDF, is a powerful technique with applications in variance reduction, biased or customized sampling, data augmentation, and beyond. Inspired by the growing availability of score-based generative models (SGMs), we propose an entirely training-free Importance sampling framework that relies solely on an SGM for the base PDF. Our key innovation is realizing the importance sampling process as a backward diffusion process, expressed in terms of the score function of the base PDF and the specified importance weight function--both readily available--eliminating the need for any additional training. We conduct a thorough analysis demonstrating the method's scalability and effectiveness across diverse datasets and tasks, including importance sampling for industrial and natural images with neural importance weight functions. The training-free aspect of our method is particularly compelling in real-world scenarios where a single base distribution underlies multiple biased sampling tasks, each requiring a different importance weight function. To the best of our knowledge our approach is the first importance sampling framework to achieve this.
2502.04649
End-to-End Learning Framework for Solving Non-Markovian Optimal Control
cs.SY cs.LG math.OC
Integer-order calculus often falls short in capturing the long-range dependencies and memory effects found in many real-world processes. Fractional calculus addresses these gaps via fractional-order integrals and derivatives, but fractional-order dynamical systems pose substantial challenges in system identification and optimal control due to the lack of standard control methodologies. In this paper, we theoretically derive the optimal control via linear quadratic regulator (LQR) for fractional-order linear time-invariant (FOLTI) systems and develop an end-to-end deep learning framework based on this theoretical foundation. Our approach establishes a rigorous mathematical model, derives analytical solutions, and incorporates deep learning to achieve data-driven optimal control of FOLTI systems. Our key contributions include: (i) proposing an innovative system identification method control strategy for FOLTI systems, (ii) developing the first end-to-end data-driven learning framework, Fractional-Order Learning for Optimal Control (FOLOC), that learns control policies from observed trajectories, and (iii) deriving a theoretical analysis of sample complexity to quantify the number of samples required for accurate optimal control in complex real-world problems. Experimental results indicate that our method accurately approximates fractional-order system behaviors without relying on Gaussian noise assumptions, pointing to promising avenues for advanced optimal control.
2502.04655
Before It's Too Late: A State Space Model for the Early Prediction of Misinformation and Disinformation Engagement
cs.CL
In today's digital age, conspiracies and information campaigns can emerge rapidly and erode social and democratic cohesion. While recent deep learning approaches have made progress in modeling engagement through language and propagation models, they struggle with irregularly sampled data and early trajectory assessment. We present IC-Mamba, a novel state space model that forecasts social media engagement by modeling interval-censored data with integrated temporal embeddings. Our model excels at predicting engagement patterns within the crucial first 15-30 minutes of posting (RMSE 0.118-0.143), enabling rapid assessment of content reach. By incorporating interval-censored modeling into the state space framework, IC-Mamba captures fine-grained temporal dynamics of engagement growth, achieving a 4.72% improvement over state-of-the-art across multiple engagement metrics (likes, shares, comments, and emojis). Our experiments demonstrate IC-Mamba's effectiveness in forecasting both post-level dynamics and broader narrative patterns (F1 0.508-0.751 for narrative-level predictions). The model maintains strong predictive performance across extended time horizons, successfully forecasting opinion-level engagement up to 28 days ahead using observation windows of 3-10 days. These capabilities enable earlier identification of potentially problematic content, providing crucial lead time for designing and implementing countermeasures. Code is available at: https://github.com/ltian678/ic-mamba. An interactive dashboard demonstrating our results is available at: https://ic-mamba.behavioral-ds.science.
2502.04656
MHAF-YOLO: Multi-Branch Heterogeneous Auxiliary Fusion YOLO for accurate object detection
cs.CV
Due to the effective multi-scale feature fusion capabilities of the Path Aggregation FPN (PAFPN), it has become a widely adopted component in YOLO-based detectors. However, PAFPN struggles to integrate high-level semantic cues with low-level spatial details, limiting its performance in real-world applications, especially with significant scale variations. In this paper, we propose MHAF-YOLO, a novel detection framework featuring a versatile neck design called the Multi-Branch Auxiliary FPN (MAFPN), which consists of two key modules: the Superficial Assisted Fusion (SAF) and Advanced Assisted Fusion (AAF). The SAF bridges the backbone and the neck by fusing shallow features, effectively transferring crucial low-level spatial information with high fidelity. Meanwhile, the AAF integrates multi-scale feature information at deeper neck layers, delivering richer gradient information to the output layer and further enhancing the model learning capacity. To complement MAFPN, we introduce the Global Heterogeneous Flexible Kernel Selection (GHFKS) mechanism and the Reparameterized Heterogeneous Multi-Scale (RepHMS) module to enhance feature fusion. RepHMS is globally integrated into the network, utilizing GHFKS to select larger convolutional kernels for various feature layers, expanding the vertical receptive field and capturing contextual information across spatial hierarchies. Locally, it optimizes convolution by processing both large and small kernels within the same layer, broadening the lateral receptive field and preserving crucial details for detecting smaller targets. The source code of this work is available at: https://github.com/yang0201/MHAF-YOLO.
2502.04658
Shifting Attention to You: Personalized Brain-Inspired AI Models
q-bio.NC cs.AI
The integration of human and artificial intelligence represents a scientific opportunity to advance our understanding of information processing, as each system offers unique computational insights that can enhance and inform the other. The synthesis of human cognitive principles with artificial intelligence has the potential to produce more interpretable and functionally aligned computational models, while simultaneously providing a formal framework for investigating the neural mechanisms underlying perception, learning, and decision-making through systematic model comparisons and representational analyses. In this study, we introduce personalized brain-inspired modeling that integrates human behavioral embeddings and neural data to align with cognitive processes. We took a stepwise approach, fine-tuning the Contrastive Language-Image Pre-training (CLIP) model with large-scale behavioral decisions, group-level neural data, and finally, participant-level neural data within a broader framework that we have named CLIP-Human-Based Analysis (CLIP-HBA). We found that fine-tuning on behavioral data enhances its ability to predict human similarity judgments while indirectly aligning it with dynamic representations captured via MEG. To further gain mechanistic insights into the temporal evolution of cognitive processes, we introduced a model specifically fine-tuned on millisecond-level MEG neural dynamics (CLIP-HBA-MEG). This model resulted in enhanced temporal alignment with human neural processing while still showing improvement on behavioral alignment. Finally, we trained individualized models on participant-specific neural data, effectively capturing individualized neural dynamics and highlighting the potential for personalized AI systems. These personalized systems have far-reaching implications for the fields of medicine, cognitive research, human-computer interfaces, and AI development.
2502.04662
Adversarially-Robust TD Learning with Markovian Data: Finite-Time Rates and Fundamental Limits
cs.LG cs.SY eess.SY math.OC
One of the most basic problems in reinforcement learning (RL) is policy evaluation: estimating the long-term return, i.e., value function, corresponding to a given fixed policy. The celebrated Temporal Difference (TD) learning algorithm addresses this problem, and recent work has investigated finite-time convergence guarantees for this algorithm and variants thereof. However, these guarantees hinge on the reward observations being always generated from a well-behaved (e.g., sub-Gaussian) true reward distribution. Motivated by harsh, real-world environments where such an idealistic assumption may no longer hold, we revisit the policy evaluation problem from the perspective of adversarial robustness. In particular, we consider a Huber-contaminated reward model where an adversary can arbitrarily corrupt each reward sample with a small probability $\epsilon$. Under this observation model, we first show that the adversary can cause the vanilla TD algorithm to converge to any arbitrary value function. We then develop a novel algorithm called Robust-TD and prove that its finite-time guarantees match that of vanilla TD with linear function approximation up to a small $O(\epsilon)$ term that captures the effect of corruption. We complement this result with a minimax lower bound, revealing that such an additive corruption-induced term is unavoidable. To our knowledge, these results are the first of their kind in the context of adversarial robustness of stochastic approximation schemes driven by Markov noise. The key new technical tool that enables our results is an analysis of the Median-of-Means estimator with corrupted, time-correlated data that might be of independent interest to the literature on robust statistics.
2502.04664
Implicit Bias of SignGD and Adam on Multiclass Separable Data
cs.LG math.OC
In the optimization of overparameterized models, different gradient-based methods can achieve zero training error yet converge to distinctly different solutions inducing different generalization properties. While a decade of research on implicit optimization bias has illuminated this phenomenon in various settings, even the foundational case of linear classification with separable data still has important open questions. We resolve a fundamental gap by characterizing the implicit bias of both Adam and Sign Gradient Descent in multi-class cross-entropy minimization: we prove that their iterates converge to solutions that maximize the margin with respect to the classifier matrix's max-norm and characterize the rate of convergence. We extend our results to general p-norm normalized steepest descent algorithms and to other multi-class losses.
2502.04666
Enhancing Health Information Retrieval with RAG by Prioritizing Topical Relevance and Factual Accuracy
cs.IR
The exponential surge in online health information, coupled with its increasing use by non-experts, highlights the pressing need for advanced Health Information Retrieval models that consider not only topical relevance but also the factual accuracy of the retrieved information, given the potential risks associated with health misinformation. To this aim, this paper introduces a solution driven by Retrieval-Augmented Generation (RAG), which leverages the capabilities of generative Large Language Models (LLMs) to enhance the retrieval of health-related documents grounded in scientific evidence. In particular, we propose a three-stage model: in the first stage, the user's query is employed to retrieve topically relevant passages with associated references from a knowledge base constituted by scientific literature. In the second stage, these passages, alongside the initial query, are processed by LLMs to generate a contextually relevant rich text (GenText). In the last stage, the documents to be retrieved are evaluated and ranked both from the point of view of topical relevance and factual accuracy by means of their comparison with GenText, either through stance detection or semantic similarity. In addition to calculating factual accuracy, GenText can offer a layer of explainability for it, aiding users in understanding the reasoning behind the retrieval. Experimental evaluation of our model on benchmark datasets and against baseline models demonstrates its effectiveness in enhancing the retrieval of both topically relevant and factually accurate health information, thus presenting a significant step forward in the health misinformation mitigation problem.
2502.04667
Unveiling the Mechanisms of Explicit CoT Training: How Chain-of-Thought Enhances Reasoning Generalization
cs.LG cs.AI cs.CL
Training large language models (LLMs) with high-quality Chain-of-Thought (CoT) annotations has become a widely adopted strategy due to its significant enhancement of reasoning capabilities. To fully comprehend this approach, two questions naturally arise: (Q1) What advantages does training with CoT offer compared to training without CoT? (Q2) If there are advantages, what are the underlying mechanisms of explicit CoT training? Analyzing the advantages and mechanisms of CoT training is challenging due to the many factors involved. To address this, we conduct a detailed analysis using clear and controllable data distributions and, for the first time, reveal that CoT training offers the following advantages: (1) Training with CoT markedly improves reasoning generalization, extending it from in-distribution (ID) to both ID and out-of-distribution (OOD) scenarios, while also speeding up convergence; (2) Even when training with CoT includes a certain range of erroneous reasoning steps, it still enables the model to learn reasoning patterns, leading to systematic generalization. We further explore the underlying mechanisms from a circuit perspective: (1) The data distribution (e.g., ratio $\lambda$ and pattern) plays a crucial role in influencing the model's systematic generalization; (2) CoT training (with two-hop facts) internalizes reasoning into a two-stage generalizing circuit, where the number of stages corresponds to the explicit reasoning steps during training. Our findings elucidate the mechanisms underlying explicit CoT training and offer critical insights into tuning strategies for LLMs to achieve robust generalization.
2502.04668
Machine-Learning Interatomic Potentials for Long-Range Systems
physics.chem-ph cs.LG
Machine-learning interatomic potentials have emerged as a revolutionary class of force-field models in molecular simulations, delivering quantum-mechanical accuracy at a fraction of the computational cost and enabling the simulation of large-scale systems over extended timescales. However, they often focus on modeling local environments, neglecting crucial long-range interactions. We propose a Sum-of-Gaussians Neural Network (SOG-Net), a lightweight and versatile framework for integrating long-range interactions into machine learning force field. The SOG-Net employs a latent-variable learning network that seamlessly bridges short-range and long-range components, coupled with an efficient Fourier convolution layer that incorporates long-range effects. By learning sum-of-Gaussian multipliers across different convolution layers, the SOG-Net adaptively captures diverse long-range decay behaviors while maintaining close-to-linear computational complexity during training and simulation via non-uniform fast Fourier transforms. The method is demonstrated effective for a broad range of long-range systems.
2502.04669
A Comprehensive Review on Noise Control of Diffusion Model
cs.LG cs.AI
Diffusion models have recently emerged as powerful generative frameworks for producing high-quality images. A pivotal component of these models is the noise schedule, which governs the rate of noise injection during the diffusion process. Since the noise schedule substantially influences sampling quality and training quality, understanding its design and implications is crucial. In this discussion, various noise schedules are examined, and their distinguishing features and performance characteristics are highlighted.
2502.04670
CCS: Controllable and Constrained Sampling with Diffusion Models via Initial Noise Perturbation
cs.LG cs.AI
Diffusion models have emerged as powerful tools for generative tasks, producing high-quality outputs across diverse domains. However, how the generated data responds to the initial noise perturbation in diffusion models remains under-explored, which hinders understanding the controllability of the sampling process. In this work, we first observe an interesting phenomenon: the relationship between the change of generation outputs and the scale of initial noise perturbation is highly linear through the diffusion ODE sampling. Then we provide both theoretical and empirical study to justify this linearity property of this input-output (noise-generation data) relationship. Inspired by these new insights, we propose a novel Controllable and Constrained Sampling method (CCS) together with a new controller algorithm for diffusion models to sample with desired statistical properties while preserving good sample quality. We perform extensive experiments to compare our proposed sampling approach with other methods on both sampling controllability and sampled data quality. Results show that our CCS method achieves more precisely controlled sampling while maintaining superior sample quality and diversity.
2502.04671
ProofWala: Multilingual Proof Data Synthesis and Theorem-Proving
cs.AI cs.LG cs.LO cs.PL
Neural networks have shown substantial promise at automatic theorem-proving in interactive proof assistants (ITPs) like Lean and Coq. However, most neural theorem-proving models are restricted to specific ITPs, leaving out opportunities for cross-lingual $\textit{transfer}$ between ITPs. We address this weakness with a multilingual proof framework, ${\rm P{\small ROOF}W{\small ALA}}$, that allows a standardized form of interaction between neural theorem-provers and two established ITPs (Coq and Lean). It enables the collection of multilingual proof step data -- data recording the result of proof actions on ITP states -- for training neural provers. ${\rm P{\small ROOF}W{\small ALA}}$ allows the systematic evaluation of a model's performance across different ITPs and problem domains via efficient parallel proof search algorithms. We show that multilingual training enabled by ${\rm P{\small ROOF}W{\small ALA}}$ can lead to successful transfer across ITPs. Specifically, a model trained on a mix of ${\rm P{\small ROOF}W{\small ALA}}$-generated Coq and Lean data outperforms Lean-only and Coq-only models on the standard prove-at-$k$ metric. We open source all code including code for the ${\rm P{\small ROOF}W{\small ALA}}$ Framework (https://github.com/trishullab/proof-wala), and the Multilingual ITP interaction framework (https://github.com/trishullab/itp-interface).
2502.04673
Optimistic Algorithms for Adaptive Estimation of the Average Treatment Effect
stat.ML cs.LG stat.ME
Estimation and inference for the Average Treatment Effect (ATE) is a cornerstone of causal inference and often serves as the foundation for developing procedures for more complicated settings. Although traditionally analyzed in a batch setting, recent advances in martingale theory have paved the way for adaptive methods that can enhance the power of downstream inference. Despite these advances, progress in understanding and developing adaptive algorithms remains in its early stages. Existing work either focus on asymptotic analyses that overlook exploration-exploitation tradeoffs relevant in finite-sample regimes or rely on simpler but suboptimal estimators. In this work, we address these limitations by studying adaptive sampling procedures that take advantage of the asymptotically optimal Augmented Inverse Probability Weighting (AIPW) estimator. Our analysis uncovers challenges obscured by asymptotic approaches and introduces a novel algorithmic design principle reminiscent of optimism in multiarmed bandits. This principled approach enables our algorithm to achieve significant theoretical and empirical gains compared to prior methods. Our findings mark a step forward in advancing adaptive causal inference methods in theory and practice.
2502.04674
AdParaphrase: Paraphrase Dataset for Analyzing Linguistic Features toward Generating Attractive Ad Texts
cs.CL cs.AI
Effective linguistic choices that attract potential customers play crucial roles in advertising success. This study aims to explore the linguistic features of ad texts that influence human preferences. Although the creation of attractive ad texts is an active area of research, progress in understanding the specific linguistic features that affect attractiveness is hindered by several obstacles. First, human preferences are complex and influenced by multiple factors, including their content, such as brand names, and their linguistic styles, making analysis challenging. Second, publicly available ad text datasets that include human preferences are lacking, such as ad performance metrics and human feedback, which reflect people's interests. To address these problems, we present AdParaphrase, a paraphrase dataset that contains human preferences for pairs of ad texts that are semantically equivalent but differ in terms of wording and style. This dataset allows for preference analysis that focuses on the differences in linguistic features. Our analysis revealed that ad texts preferred by human judges have higher fluency, longer length, more nouns, and use of bracket symbols. Furthermore, we demonstrate that an ad text-generation model that considers these findings significantly improves the attractiveness of a given text. The dataset is publicly available at: https://github.com/CyberAgentAILab/AdParaphrase.
2502.04675
Scalable Oversight for Superhuman AI via Recursive Self-Critiquing
cs.AI cs.CL cs.LG
As AI capabilities increasingly surpass human proficiency in complex tasks, current alignment techniques including SFT and RLHF face fundamental challenges in ensuring reliable oversight. These methods rely on direct human assessment and become untenable when AI outputs exceed human cognitive thresholds. In response to this challenge, we explore two hypotheses: (1) critique of critique can be easier than critique itself, extending the widely-accepted observation that verification is easier than generation to the critique domain, as critique itself is a specialized form of generation; (2) this difficulty relationship is recursively held, suggesting that when direct evaluation is infeasible, performing high-order critiques (e.g., critique of critique of critique) offers a more tractable supervision pathway. To examine these hypotheses, we perform Human-Human, Human-AI, and AI-AI experiments across multiple tasks. Our results demonstrate encouraging evidence supporting these hypotheses and suggest that recursive self-critiquing is a promising direction for scalable oversight.
2502.04678
Nearly Tight Bounds for Cross-Learning Contextual Bandits with Graphical Feedback
cs.LG
The cross-learning contextual bandit problem with graphical feedback has recently attracted significant attention. In this setting, there is a contextual bandit with a feedback graph over the arms, and pulling an arm reveals the loss for all neighboring arms in the feedback graph across all contexts. Initially proposed by Han et al. (2024), this problem has broad applications in areas such as bidding in first price auctions, and explores a novel frontier in the feedback structure of bandit problems. A key theoretical question is whether an algorithm with $\widetilde{O}(\sqrt{\alpha T})$ regret exists, where $\alpha$ represents the independence number of the feedback graph. This question is particularly interesting because it concerns whether an algorithm can achieve a regret bound entirely independent of the number of contexts and matching the minimax regret of vanilla graphical bandits. Previous work has demonstrated that such an algorithm is impossible for adversarial contexts, but the question remains open for stochastic contexts. In this work, we affirmatively answer this open question by presenting an algorithm that achieves the minimax $\widetilde{O}(\sqrt{\alpha T})$ regret for cross-learning contextual bandits with graphical feedback and stochastic contexts. Notably, although that question is open even for stochastic bandits, we directly solve the strictly stronger adversarial bandit version of the problem.
2502.04679
Mechanistic Understandings of Representation Vulnerabilities and Engineering Robust Vision Transformers
cs.CV cs.LG
While transformer-based models dominate NLP and vision applications, their underlying mechanisms to map the input space to the label space semantically are not well understood. In this paper, we study the sources of known representation vulnerabilities of vision transformers (ViT), where perceptually identical images can have very different representations and semantically unrelated images can have the same representation. Our analysis indicates that imperceptible changes to the input can result in significant representation changes, particularly in later layers, suggesting potential instabilities in the performance of ViTs. Our comprehensive study reveals that adversarial effects, while subtle in early layers, propagate and amplify through the network, becoming most pronounced in middle to late layers. This insight motivates the development of NeuroShield-ViT, a novel defense mechanism that strategically neutralizes vulnerable neurons in earlier layers to prevent the cascade of adversarial effects. We demonstrate NeuroShield-ViT's effectiveness across various attacks, particularly excelling against strong iterative attacks, and showcase its remarkable zero-shot generalization capabilities. Without fine-tuning, our method achieves a competitive accuracy of 77.8% on adversarial examples, surpassing conventional robustness methods. Our results shed new light on how adversarial effects propagate through ViT layers, while providing a promising approach to enhance the robustness of vision transformers against adversarial attacks. Additionally, they provide a promising approach to enhance the robustness of vision transformers against adversarial attacks.
2502.04680
Performance Evaluation of Image Enhancement Techniques on Transfer Learning for Touchless Fingerprint Recognition
cs.CV cs.LG
Fingerprint recognition remains one of the most reliable biometric technologies due to its high accuracy and uniqueness. Traditional systems rely on contact-based scanners, which are prone to issues such as image degradation from surface contamination and inconsistent user interaction. To address these limitations, contactless fingerprint recognition has emerged as a promising alternative, providing non-intrusive and hygienic authentication. This study evaluates the impact of image enhancement tech-niques on the performance of pre-trained deep learning models using transfer learning for touchless fingerprint recognition. The IIT-Bombay Touchless and Touch-Based Fingerprint Database, containing data from 200 subjects, was employed to test the per-formance of deep learning architectures such as VGG-16, VGG-19, Inception-V3, and ResNet-50. Experimental results reveal that transfer learning methods with fingerprint image enhance-ment (indirect method) significantly outperform those without enhancement (direct method). Specifically, VGG-16 achieved an accuracy of 98% in training and 93% in testing when using the enhanced images, demonstrating superior performance compared to the direct method. This paper provides a detailed comparison of the effectiveness of image enhancement in improving the accuracy of transfer learning models for touchless fingerprint recognition, offering key insights for developing more efficient biometric systems.
2502.04682
AI-Driven Solutions for Falcon Disease Classification: Concatenated ConvNeXt cum EfficientNet AI Model Approach
cs.CV
Falconry, an ancient practice of training and hunting with falcons, emphasizes the need for vigilant health monitoring to ensure the well-being of these highly valued birds, especially during hunting activities. This research paper introduces a cutting-edge approach, which leverages the power of Concatenated ConvNeXt and EfficientNet AI models for falcon disease classification. Focused on distinguishing 'Normal,' 'Liver,' and 'Aspergillosis' cases, the study employs a comprehensive dataset for model training and evaluation, utilizing metrics such as accuracy, precision, recall, and f1-score. Through rigorous experimentation and evaluation, we demonstrate the superior performance of the concatenated AI model compared to traditional methods and standalone architectures. This novel approach contributes to accurate falcon disease classification, laying the groundwork for further advancements in avian veterinary AI applications.
2502.04684
G2PDiffusion: Genotype-to-Phenotype Prediction with Diffusion Models
cs.LG cs.AI
Discovering the genotype-phenotype relationship is crucial for genetic engineering, which will facilitate advances in fields such as crop breeding, conservation biology, and personalized medicine. Current research usually focuses on single species and small datasets due to limitations in phenotypic data collection, especially for traits that require visual assessments or physical measurements. Deciphering complex and composite phenotypes, such as morphology, from genetic data at scale remains an open question. To break through traditional generic models that rely on simplified assumptions, this paper introduces G2PDiffusion, the first-of-its-kind diffusion model designed for genotype-to-phenotype generation across multiple species. Specifically, we use images to represent morphological phenotypes across species and redefine phenotype prediction as conditional image generation. To this end, this paper introduces an environment-enhanced DNA sequence conditioner and trains a stable diffusion model with a novel alignment method to improve genotype-to-phenotype consistency. Extensive experiments demonstrate that our approach enhances phenotype prediction accuracy across species, capturing subtle genetic variations that contribute to observable traits.
2502.04686
Learning Strategic Language Agents in the Werewolf Game with Iterative Latent Space Policy Optimization
cs.AI
Large language model (LLM)-based agents have recently shown impressive progress in a variety of domains, including open-ended conversation and multi-step decision-making. However, applying these agents to social deduction games such as Werewolf, which requires both strategic decision-making and free-form language interaction, remains non-trivial. Traditional methods based on Counterfactual Regret Minimization (CFR) or reinforcement learning (RL) typically depend on a predefined action space, making them unsuitable for language games with unconstrained text action space. Meanwhile, pure LLM-based agents often suffer from intrinsic biases and require prohibitively large datasets for fine-tuning. We propose Latent Space Policy Optimization (LSPO), an iterative framework that addresses these challenges by first mapping free-form text to a discrete latent space, where methods like CFR and RL can learn strategic policy more effectively. We then translate the learned policy back into natural language dialogues, which are used to fine-tune an LLM via Direct Preference Optimization (DPO). By iteratively alternating between these stages, our LSPO agent progressively enhances both strategic reasoning and language communication. Experiment results on the Werewolf game show that our method improves the agent's performance in each iteration and outperforms existing Werewolf agents, underscoring its promise for free-form language decision-making.
2502.04688
M-IFEval: Multilingual Instruction-Following Evaluation
cs.CL cs.AI
Instruction following is a core capability of modern Large language models (LLMs), making evaluating this capability essential to understanding these models. The Instruction Following Evaluation (IFEval) benchmark from the literature does this using objective criteria, offering a measure of LLM performance without subjective AI or human judgement. However, it only includes English instructions, limiting its ability to assess LLMs in other languages. We propose the Multilingual Instruction Following Evaluation (M-IFEval) benchmark, expanding the evaluation to French, Japanese, and Spanish, with both general and language-specific instructions. Applying this benchmark to 8 state-of-the-art LLMs, we find that benchmark performance across languages and instruction types can vary widely, underscoring the importance of a multilingual benchmark for evaluating LLMs in a diverse cultural context.
2502.04689
ARR: Question Answering with Large Language Models via Analyzing, Retrieving, and Reasoning
cs.CL cs.AI cs.LG
Large language models (LLMs) achieve remarkable performance on challenging benchmarks that are often structured as multiple-choice question-answering (QA) tasks. Zero-shot Chain-of-Thought (CoT) prompting enhances reasoning in LLMs but provides only vague and generic guidance ("think step by step"). This paper introduces ARR, an intuitive and effective zero-shot prompting method that explicitly incorporates three key steps in QA solving: analyzing the intent of the question, retrieving relevant information, and reasoning step by step. Comprehensive experiments across diverse and challenging QA tasks demonstrate that ARR consistently improves the Baseline (without ARR prompting) and outperforms CoT. Ablation and case studies further validate the positive contributions of each component: analyzing, retrieving, and reasoning. Notably, intent analysis plays a vital role in ARR. Additionally, extensive evaluations across various model sizes, LLM series, and generation settings solidify the effectiveness, robustness, and generalizability of ARR.
2502.04692
STRIDE: Automating Reward Design, Deep Reinforcement Learning Training and Feedback Optimization in Humanoid Robotics Locomotion
cs.RO cs.LG
Humanoid robotics presents significant challenges in artificial intelligence, requiring precise coordination and control of high-degree-of-freedom systems. Designing effective reward functions for deep reinforcement learning (DRL) in this domain remains a critical bottleneck, demanding extensive manual effort, domain expertise, and iterative refinement. To overcome these challenges, we introduce STRIDE, a novel framework built on agentic engineering to automate reward design, DRL training, and feedback optimization for humanoid robot locomotion tasks. By combining the structured principles of agentic engineering with large language models (LLMs) for code-writing, zero-shot generation, and in-context optimization, STRIDE generates, evaluates, and iteratively refines reward functions without relying on task-specific prompts or templates. Across diverse environments featuring humanoid robot morphologies, STRIDE outperforms the state-of-the-art reward design framework EUREKA, achieving an average improvement of round 250% in efficiency and task performance. Using STRIDE-generated rewards, simulated humanoid robots achieve sprint-level locomotion across complex terrains, highlighting its ability to advance DRL workflows and humanoid robotics research.
2502.04693
Be Water, My Antennas: Riding on Radio Wave Fluctuation in Nature for Spatial Multiplexing using Programmable Meta-Fluid Antenna
physics.app-ph cs.SY eess.SY
Interference and scattering, often deemed undesirable, are inevitable in wireless communications, especially when the current mobile networks and upcoming sixth generation (6G) have turned into ultra-dense networks. Current approaches relying on multiple-input multiple-output (MIMO) combined with artificial-intelligence-aided (AI) signal processing have drawbacks of being power-hungry and requiring wide bandwidth that raise scalability concerns. In this article, we take a radical approach and utilize the channel fading phenomenon to our advantage. Specifically, we propose a novel meta-fluid antenna architecture, referred to as the `fluid' antenna system (FAS), that can freely surf on radio wave fluctuations, like `fluid' figuratively speaking, with fine resolution in space to opportunistically avoid interference, eliminating the need for expensive signal processing. Our experimental results demonstrate that under rich scattering conditions, the proposed meta-fluidic architecture is able to exploit the natural ups and downs of radio waves in space for spatial multiplexing. These breakthrough results show that scattering can be desirable not harmful and interference can be dodged not suppressed, fundamentally changing our perception of fading and our understanding on how interference should be managed in wireless communications networks.
2502.04695
Bridging the Gap in XAI-Why Reliable Metrics Matter for Explainability and Compliance
cs.AI cs.CE cs.ET cs.LG
This position paper emphasizes the critical gap in the evaluation of Explainable AI (XAI) due to the lack of standardized and reliable metrics, which diminishes its practical value, trustworthiness, and ability to meet regulatory requirements. Current evaluation methods are often fragmented, subjective, and biased, making them prone to manipulation and complicating the assessment of complex models. A central issue is the absence of a ground truth for explanations, complicating comparisons across various XAI approaches. To address these challenges, we advocate for widespread research into developing robust, context-sensitive evaluation metrics. These metrics should be resistant to manipulation, relevant to each use case, and based on human judgment and real-world applicability. We also recommend creating domain-specific evaluation benchmarks that align with the user and regulatory needs of sectors such as healthcare and finance. By encouraging collaboration among academia, industry, and regulators, we can create standards that balance flexibility and consistency, ensuring XAI explanations are meaningful, trustworthy, and compliant with evolving regulations.
2502.04696
Adaptive Learning-based Model Predictive Control Strategy for Drift Vehicles
cs.RO
Drift vehicle control offers valuable insights to support safe autonomous driving in extreme conditions, which hinges on tracking a particular path while maintaining the vehicle states near the drift equilibrium points (DEP). However, conventional tracking methods are not adaptable for drift vehicles due to their opposite steering angle and yaw rate. In this paper, we propose an adaptive path tracking (APT) control method to dynamically adjust drift states to follow the reference path, improving the commonly utilized predictive path tracking methods with released computation burden. Furthermore, existing control strategies necessitate a precise system model to calculate the DEP, which can be more intractable due to the highly nonlinear drift dynamics and sensitive vehicle parameters. To tackle this problem, an adaptive learning-based model predictive control (ALMPC) strategy is proposed based on the APT method, where an upper-level Bayesian optimization is employed to learn the DEP and APT control law to instruct a lower-level MPC drift controller. This hierarchical system architecture can also resolve the inherent control conflict between path tracking and drifting by separating these objectives into different layers. The ALMPC strategy is verified on the Matlab-Carsim platform, and simulation results demonstrate its effectiveness in controlling the drift vehicle to follow a clothoid-based reference path even with the misidentified road friction parameter.
2502.04697
Multi-Agent Coverage Control in Non-Convex Annulus Region with Conformal Mapping
eess.SY cs.SY
Efficiently fulfilling coverage tasks in non-convex regions has long been a significant challenge for multi-agent systems (MASs). By leveraging conformal mapping, this paper introduces a novel sectorial coverage formulation to transform a non-convex annulus region into a topologically equivalent one. This approach enables the deployment of MASs in a non-star-shaped region while optimizing coverage performance and achieving load balance among sub-regions. It provides a unique perspective on the partitioned sub-regions to highlight the geodesic convex property of the non-star-shaped region. By utilizing the sectorial partition mechanism and the diffeomorphism property of conformal mapping, a decentralized control law is designed to drive MASs towards a desired configuration, which not only optimizes the global coverage cost but also ensures exponential convergence of equitable workload. Moreover, an iterative search algorithm is developed to identify the optimal approximation of multi-agent deployment in the non-star-shaped region. Theoretical analysis is conducted to confirm the asymptotic stability and global convergence with arbitrary small tolerance of the closed-loop system. Finally, numerical simulations demonstrate the practicality of the proposed coverage formulation with conformal mapping.
2502.04699
A Meta-learner for Heterogeneous Effects in Difference-in-Differences
stat.ML cs.LG
We address the problem of estimating heterogeneous treatment effects in panel data, adopting the popular Difference-in-Differences (DiD) framework under the conditional parallel trends assumption. We propose a novel doubly robust meta-learner for the Conditional Average Treatment Effect on the Treated (CATT), reducing the estimation to a convex risk minimization problem involving a set of auxiliary models. Our framework allows for the flexible estimation of the CATT, when conditioning on any subset of variables of interest using generic machine learning. Leveraging Neyman orthogonality, our proposed approach is robust to estimation errors in the auxiliary models. As a generalization to our main result, we develop a meta-learning approach for the estimation of general conditional functionals under covariate shift. We also provide an extension to the instrumented DiD setting with non-compliance. Empirical results demonstrate the superiority of our approach over existing baselines.
2502.04700
EigenLoRAx: Recycling Adapters to Find Principal Subspaces for Resource-Efficient Adaptation and Inference
cs.LG cs.AI
The rapid growth of large models has raised concerns about their environmental impact and equity in accessibility due to significant computational costs. Low-Rank Adapters (LoRA) offer a lightweight solution for finetuning large models, resulting in an abundance of publicly available adapters tailored to diverse domains. We ask: Can these pretrained adapters be leveraged to further streamline adaptation to new tasks while addressing these challenges? We introduce EigenLoRAx, a parameter-efficient finetuning method that recycles existing adapters to create a principal subspace aligned with their shared domain knowledge which can be further augmented with orthogonal basis vectors in low-resource scenarios. This enables rapid adaptation to new tasks by learning only lightweight coefficients on the principal components of the subspace - eliminating the need to finetune entire adapters. EigenLoRAx requires significantly fewer parameters and memory, improving efficiency for both training and inference. Our method demonstrates strong performance across diverse domains and tasks, offering a scalable for edge-based applications, personalization, and equitable deployment of large models in resource-constrained environments.
2502.04703
Symbolic Regression of Data-Driven Reduced Order Model Closures for Under-Resolved, Convection-Dominated Flows
math.NA cs.LG cs.NA physics.flu-dyn
Data-driven closures correct the standard reduced order models (ROMs) to increase their accuracy in under-resolved, convection-dominated flows. There are two types of data-driven ROM closures in current use: (i) structural, with simple ansatzes (e.g., linear or quadratic); and (ii) machine learning-based, with neural network ansatzes. We propose a novel symbolic regression (SR) data-driven ROM closure strategy, which combines the advantages of current approaches and eliminates their drawbacks. As a result, the new data-driven SR closures yield ROMs that are interpretable, parsimonious, accurate, generalizable, and robust. To compare the data-driven SR-ROM closures with the structural and machine learning-based ROM closures, we consider the data-driven variational multiscale ROM framework and two under-resolved, convection-dominated test problems: the flow past a cylinder and the lid-driven cavity flow at Reynolds numbers Re = 10000, 15000, and 20000. This numerical investigation shows that the new data-driven SR-ROM closures yield more accurate and robust ROMs than the structural and machine learning ROM closures.
2502.04706
Enhancing Impression Change Prediction in Speed Dating Simulations Based on Speakers' Personalities
cs.CL cs.HC
This paper focuses on simulating text dialogues in which impressions between speakers improve during speed dating. This simulation involves selecting an utterance from multiple candidates generated by a text generation model that replicates a specific speaker's utterances, aiming to improve the impression of the speaker. Accurately selecting an utterance that improves the impression is crucial for the simulation. We believe that whether an utterance improves a dialogue partner's impression of the speaker may depend on the personalities of both parties. However, recent methods for utterance selection do not consider the impression per utterance or the personalities. To address this, we propose a method that predicts whether an utterance improves a partner's impression of the speaker, considering the personalities. The evaluation results showed that personalities are useful in predicting impression changes per utterance. Furthermore, we conducted a human evaluation of simulated dialogues using our method. The results showed that it could simulate dialogues more favorably received than those selected without considering personalities.
2502.04713
Leveraging band diversity for feature selection in EO data
eess.IV cs.CV
Hyperspectral imaging (HSI) is a powerful earth observation technology that captures and processes information across a wide spectrum of wavelengths. Hyperspectral imaging provides comprehensive and detailed spectral data that is invaluable for a wide range of reconstruction problems. However due to complexity in analysis it often becomes difficult to handle this data. To address the challenge of handling large number of bands in reconstructing high quality HSI, we propose to form groups of bands. In this position paper we propose a method of selecting diverse bands using determinantal point processes in correlated bands. To address the issue of overlapping bands that may arise from grouping, we use spectral angle mapper analysis. This analysis can be fed to any Machine learning model to enable detailed analysis and monitoring with high precision and accuracy.
2502.04718
Evaluating Text Style Transfer Evaluation: Are There Any Reliable Metrics?
cs.CL
Text Style Transfer (TST) is the task of transforming a text to reflect a particular style while preserving its original content. Evaluating TST outputs is a multidimensional challenge, requiring the assessment of style transfer accuracy, content preservation, and naturalness. Using human evaluation is ideal but costly, same as in other natural language processing (NLP) tasks, however, automatic metrics for TST have not received as much attention as metrics for, e.g., machine translation or summarization. In this paper, we examine both set of existing and novel metrics from broader NLP tasks for TST evaluation, focusing on two popular subtasks-sentiment transfer and detoxification-in a multilingual context comprising English, Hindi, and Bengali. By conducting meta-evaluation through correlation with human judgments, we demonstrate the effectiveness of these metrics when used individually and in ensembles. Additionally, we investigate the potential of Large Language Models (LLMs) as tools for TST evaluation. Our findings highlight that certain advanced NLP metrics and experimental-hybrid-techniques, provide better insights than existing TST metrics for delivering more accurate, consistent, and reproducible TST evaluations.
2502.04719
Tolerance-Aware Deep Optics
cs.CV cs.GR
Deep optics has emerged as a promising approach by co-designing optical elements with deep learning algorithms. However, current research typically overlooks the analysis and optimization of manufacturing and assembly tolerances. This oversight creates a significant performance gap between designed and fabricated optical systems. To address this challenge, we present the first end-to-end tolerance-aware optimization framework that incorporates multiple tolerance types into the deep optics design pipeline. Our method combines physics-informed modelling with data-driven training to enhance optical design by accounting for and compensating for structural deviations in manufacturing and assembly. We validate our approach through computational imaging applications, demonstrating results in both simulations and real-world experiments. We further examine how our proposed solution improves the robustness of optical systems and vision algorithms against tolerances through qualitative and quantitative analyses. Code and additional visual results are available at openimaginglab.github.io/LensTolerance.
2502.04722
Singing Voice Conversion with Accompaniment Using Self-Supervised Representation-Based Melody Features
cs.SD cs.LG eess.AS
Melody preservation is crucial in singing voice conversion (SVC). However, in many scenarios, audio is often accompanied with background music (BGM), which can cause audio distortion and interfere with the extraction of melody and other key features, significantly degrading SVC performance. Previous methods have attempted to address this by using more robust neural network-based melody extractors, but their performance drops sharply in the presence of complex accompaniment. Other approaches involve performing source separation before conversion, but this often introduces noticeable artifacts, leading to a significant drop in conversion quality and increasing the user's operational costs. To address these issues, we introduce a novel SVC method that uses self-supervised representation-based melody features to improve melody modeling accuracy in the presence of BGM. In our experiments, we compare the effectiveness of different self-supervised learning (SSL) models for melody extraction and explore for the first time how SSL benefits the task of melody extraction. The experimental results demonstrate that our proposed SVC model significantly outperforms existing baseline methods in terms of melody accuracy and shows higher similarity and naturalness in both subjective and objective evaluations across noisy and clean audio environments.
2502.04725
Can Diffusion Models Learn Hidden Inter-Feature Rules Behind Images?
cs.CV cs.AI
Despite the remarkable success of diffusion models (DMs) in data generation, they exhibit specific failure cases with unsatisfactory outputs. We focus on one such limitation: the ability of DMs to learn hidden rules between image features. Specifically, for image data with dependent features ($\mathbf{x}$) and ($\mathbf{y}$) (e.g., the height of the sun ($\mathbf{x}$) and the length of the shadow ($\mathbf{y}$)), we investigate whether DMs can accurately capture the inter-feature rule ($p(\mathbf{y}|\mathbf{x})$). Empirical evaluations on mainstream DMs (e.g., Stable Diffusion 3.5) reveal consistent failures, such as inconsistent lighting-shadow relationships and mismatched object-mirror reflections. Inspired by these findings, we design four synthetic tasks with strongly correlated features to assess DMs' rule-learning abilities. Extensive experiments show that while DMs can identify coarse-grained rules, they struggle with fine-grained ones. Our theoretical analysis demonstrates that DMs trained via denoising score matching (DSM) exhibit constant errors in learning hidden rules, as the DSM objective is not compatible with rule conformity. To mitigate this, we introduce a common technique - incorporating additional classifier guidance during sampling, which achieves (limited) improvements. Our analysis reveals that the subtle signals of fine-grained rules are challenging for the classifier to capture, providing insights for future exploration.
2502.04728
Generating Symbolic World Models via Test-time Scaling of Large Language Models
cs.AI
Solving complex planning problems requires Large Language Models (LLMs) to explicitly model the state transition to avoid rule violations, comply with constraints, and ensure optimality-a task hindered by the inherent ambiguity of natural language. To overcome such ambiguity, Planning Domain Definition Language (PDDL) is leveraged as a planning abstraction that enables precise and formal state descriptions. With PDDL, we can generate a symbolic world model where classic searching algorithms, such as A*, can be seamlessly applied to find optimal plans. However, directly generating PDDL domains with current LLMs remains an open challenge due to the lack of PDDL training data. To address this challenge, we propose to scale up the test-time computation of LLMs to enhance their PDDL reasoning capabilities, thereby enabling the generation of high-quality PDDL domains. Specifically, we introduce a simple yet effective algorithm, which first employs a Best-of-N sampling approach to improve the quality of the initial solution and then refines the solution in a fine-grained manner with verbalized machine learning. Our method outperforms o1-mini by a considerable margin in the generation of PDDL domain, achieving over 50% success rate on two tasks (i.e., generating PDDL domains from natural language description or PDDL problems). This is done without requiring additional training. By taking advantage of PDDL as state abstraction, our method is able to outperform current state-of-the-art methods on almost all competition-level planning tasks.
2502.04729
The "negative end" of change in grammar: terminology, concepts and causes
cs.CL cs.CY
The topic of "negative end" of change is, contrary to the fields of innovation and emergence, largely under-researched. Yet, it has lately started to gain an increasing attention from language scholars worldwide. The main focus of this article is threefold, namely to discuss the i) terminology; ii) concepts and iii) causes associated with the "negative end" of change in grammar. The article starts with an overview of research conducted on the topic. It then moves to situating phenomena referred to as loss, decline or obsolescence among processes of language change, before elaborating on the terminology and concepts behind it. The last part looks at possible causes for constructions to display a (gradual or rapid, but very consistent) decrease in the frequency of use over time, which continues until the construction disappears or there are only residual or fossilised forms left. Keywords: loss, obsolescence, decline, competition, higher
2502.04730
PhyloVAE: Unsupervised Learning of Phylogenetic Trees via Variational Autoencoders
stat.ML cs.LG q-bio.PE
Learning informative representations of phylogenetic tree structures is essential for analyzing evolutionary relationships. Classical distance-based methods have been widely used to project phylogenetic trees into Euclidean space, but they are often sensitive to the choice of distance metric and may lack sufficient resolution. In this paper, we introduce phylogenetic variational autoencoders (PhyloVAEs), an unsupervised learning framework designed for representation learning and generative modeling of tree topologies. Leveraging an efficient encoding mechanism inspired by autoregressive tree topology generation, we develop a deep latent-variable generative model that facilitates fast, parallelized topology generation. PhyloVAE combines this generative model with a collaborative inference model based on learnable topological features, allowing for high-resolution representations of phylogenetic tree samples. Extensive experiments demonstrate PhyloVAE's robust representation learning capabilities and fast generation of phylogenetic tree topologies.
2502.04734
SC-OmniGS: Self-Calibrating Omnidirectional Gaussian Splatting
cs.CV cs.GR
360-degree cameras streamline data collection for radiance field 3D reconstruction by capturing comprehensive scene data. However, traditional radiance field methods do not address the specific challenges inherent to 360-degree images. We present SC-OmniGS, a novel self-calibrating omnidirectional Gaussian splatting system for fast and accurate omnidirectional radiance field reconstruction using 360-degree images. Rather than converting 360-degree images to cube maps and performing perspective image calibration, we treat 360-degree images as a whole sphere and derive a mathematical framework that enables direct omnidirectional camera pose calibration accompanied by 3D Gaussians optimization. Furthermore, we introduce a differentiable omnidirectional camera model in order to rectify the distortion of real-world data for performance enhancement. Overall, the omnidirectional camera intrinsic model, extrinsic poses, and 3D Gaussians are jointly optimized by minimizing weighted spherical photometric loss. Extensive experiments have demonstrated that our proposed SC-OmniGS is able to recover a high-quality radiance field from noisy camera poses or even no pose prior in challenging scenarios characterized by wide baselines and non-object-centric configurations. The noticeable performance gain in the real-world dataset captured by consumer-grade omnidirectional cameras verifies the effectiveness of our general omnidirectional camera model in reducing the distortion of 360-degree images.
2502.04737
Learning Universal Multi-level Market Irrationality Factors to Improve Stock Return Forecasting
cs.LG
Recent years have witnessed the perfect encounter of deep learning and quantitative trading has achieved great success in stock investment. Numerous deep learning-based models have been developed for forecasting stock returns, leveraging the powerful representation capabilities of neural networks to identify patterns and factors influencing stock prices. These models can effectively capture general patterns in the market, such as stock price trends, volume-price relationships, and time variations. However, the impact of special irrationality factors -- such as market sentiment, speculative behavior, market manipulation, and psychological biases -- have not been fully considered in existing deep stock forecasting models due to their relative abstraction as well as lack of explicit labels and data description. To fill this gap, we propose UMI, a Universal multi-level Market Irrationality factor model to enhance stock return forecasting. The UMI model learns factors that can reflect irrational behaviors in market from both individual stock and overall market levels. For the stock-level, UMI construct an estimated rational price for each stock, which is cointegrated with the stock's actual price. The discrepancy between the actual and the rational prices serves as a factor to indicate stock-level irrational events. Additionally, we define market-level irrational behaviors as anomalous synchronous fluctuations of stocks within a market. Using two self-supervised representation learning tasks, i.e., sub-market comparative learning and market synchronism prediction, the UMI model incorporates market-level irrationalities into a market representation vector, which is then used as the market-level irrationality factor.
2502.04740
SelaFD:Seamless Adaptation of Vision Transformer Fine-tuning for Radar-based Human Activity
cs.CV cs.LG
Human Activity Recognition (HAR) such as fall detection has become increasingly critical due to the aging population, necessitating effective monitoring systems to prevent serious injuries and fatalities associated with falls. This study focuses on fine-tuning the Vision Transformer (ViT) model specifically for HAR using radar-based Time-Doppler signatures. Unlike traditional image datasets, these signals present unique challenges due to their non-visual nature and the high degree of similarity among various activities. Directly fine-tuning the ViT with all parameters proves suboptimal for this application. To address this challenge, we propose a novel approach that employs Low-Rank Adaptation (LoRA) fine-tuning in the weight space to facilitate knowledge transfer from pre-trained ViT models. Additionally, to extract fine-grained features, we enhance feature representation through the integration of a serial-parallel adapter in the feature space. Our innovative joint fine-tuning method, tailored for radar-based Time-Doppler signatures, significantly improves HAR accuracy, surpassing existing state-of-the-art methodologies in this domain. Our code is released at https://github.com/wangyijunlyy/SelaFD.