id
stringlengths
9
16
title
stringlengths
4
278
categories
stringlengths
5
104
abstract
stringlengths
6
4.09k
2502.01701
Learning with Differentially Private (Sliced) Wasserstein Gradients
cs.LG math.ST stat.TH
In this work, we introduce a novel framework for privately optimizing objectives that rely on Wasserstein distances between data-dependent empirical measures. Our main theoretical contribution is, based on an explicit formulation of the Wasserstein gradient in a fully discrete setting, a control on the sensitivity of this gradient to individual data points, allowing strong privacy guarantees at minimal utility cost. Building on these insights, we develop a deep learning approach that incorporates gradient and activations clipping, originally designed for DP training of problems with a finite-sum structure. We further demonstrate that privacy accounting methods extend to Wasserstein-based objectives, facilitating large-scale private training. Empirical results confirm that our framework effectively balances accuracy and privacy, offering a theoretically sound solution for privacy-preserving machine learning tasks relying on optimal transport distances such as Wasserstein distance or sliced-Wasserstein distance.
2502.01702
Al-Khwarizmi: Discovering Physical Laws with Foundation Models
cs.LG
Inferring physical laws from data is a central challenge in science and engineering, including but not limited to healthcare, physical sciences, biosciences, social sciences, sustainability, climate, and robotics. Deep networks offer high-accuracy results but lack interpretability, prompting interest in models built from simple components. The Sparse Identification of Nonlinear Dynamics (SINDy) method has become the go-to approach for building such modular and interpretable models. SINDy leverages sparse regression with L1 regularization to identify key terms from a library of candidate functions. However, SINDy's choice of candidate library and optimization method requires significant technical expertise, limiting its widespread applicability. This work introduces Al-Khwarizmi, a novel agentic framework for physical law discovery from data, which integrates foundational models with SINDy. Leveraging LLMs, VLMs, and Retrieval-Augmented Generation (RAG), our approach automates physical law discovery, incorporating prior knowledge and iteratively refining candidate solutions via reflection. Al-Khwarizmi operates in two steps: it summarizes system observations-comprising textual descriptions, raw data, and plots-followed by a secondary step that generates candidate feature libraries and optimizer configurations to identify hidden physics laws correctly. Evaluating our algorithm on over 198 models, we demonstrate state-of-the-art performance compared to alternatives, reaching a 20 percent increase against the best-performing alternative.
2502.01703
QLESS: A Quantized Approach for Data Valuation and Selection in Large Language Model Fine-Tuning
cs.LG cs.AI cs.CL
Fine-tuning large language models (LLMs) is often constrained by the computational costs of processing massive datasets. We propose \textbf{QLESS} (Quantized Low-rank Gradient Similarity Search), which integrates gradient quantization with the LESS framework to enable memory-efficient data valuation and selection. QLESS employs a two-step compression process: first, it obtains low-dimensional gradient representations through LoRA-based random projection; then, it quantizes these gradients to low-bitwidth representations. Experiments on multiple LLM architectures (LLaMA, Mistral, Qwen) and benchmarks (MMLU, BBH, TyDiQA) show that QLESS achieves comparable data selection performance to LESS while reducing memory usage by up to 16x. Even 1-bit gradient quantization preserves data valuation quality. These findings underscore QLESS as a practical, scalable approach to identifying informative examples within strict memory constraints.
2502.01704
Adaptive Observation Cost Control for Variational Quantum Eigensolvers
quant-ph cs.LG
The objective to be minimized in the variational quantum eigensolver (VQE) has a restricted form, which allows a specialized sequential minimal optimization (SMO) that requires only a few observations in each iteration. However, the SMO iteration is still costly due to the observation noise -- one observation at a point typically requires averaging over hundreds to thousands of repeated quantum measurement shots for achieving a reasonable noise level. In this paper, we propose an adaptive cost control method, named subspace in confident region (SubsCoRe), for SMO. SubsCoRe uses the Gaussian process (GP) surrogate, and requires it to have low uncertainty over the subspace being updated, so that optimization in each iteration is performed with guaranteed accuracy. The adaptive cost control is performed by first setting the required accuracy according to the progress of the optimization, and then choosing the minimum number of measurement shots and their distribution such that the required accuracy is satisfied. We demonstrate that SubsCoRe significantly improves the efficiency of SMO, and outperforms the state-of-the-art methods.
2502.01705
Progressive Binarization with Semi-Structured Pruning for LLMs
cs.LG
Large language models (LLMs) have achieved remarkable success in natural language processing tasks, but their high computational and memory demands pose challenges for deployment on resource-constrained devices. Binarization, as an efficient compression method that reduces model weights to just 1 bit, significantly lowers both computational and memory requirements. Despite this, the binarized LLM still contains redundancy, which can be further compressed. Semi-structured pruning provides a promising approach to achieve this, which offers a better trade-off between model performance and hardware efficiency. However, simply combining binarization with semi-structured pruning can lead to a significant performance drop. To address this issue, we propose a Progressive Binarization with Semi-Structured Pruning (PBS$^2$P) method for LLM compression. We first propose a Stepwise semi-structured Pruning with Binarization Optimization (SPBO). Our optimization strategy significantly reduces the total error caused by pruning and binarization, even below that of the no-pruning scenario. Furthermore, we design a Coarse-to-Fine Search (CFS) method to select pruning elements more effectively. Extensive experiments demonstrate that PBS$^2$P achieves superior accuracy across various LLM families and evaluation metrics, noticeably outperforming state-of-the-art (SOTA) binary PTQ methods. The code and models will be available at https://github.com/XIANGLONGYAN/PBS2P.
2502.01706
Comply: Learning Sentences with Complex Weights inspired by Fruit Fly Olfaction
cs.CL cs.AI cs.LG cs.NE
Biologically inspired neural networks offer alternative avenues to model data distributions. FlyVec is a recent example that draws inspiration from the fruit fly's olfactory circuit to tackle the task of learning word embeddings. Surprisingly, this model performs competitively even against deep learning approaches specifically designed to encode text, and it does so with the highest degree of computational efficiency. We pose the question of whether this performance can be improved further. For this, we introduce Comply. By incorporating positional information through complex weights, we enable a single-layer neural network to learn sequence representations. Our experiments show that Comply not only supersedes FlyVec but also performs on par with significantly larger state-of-the-art models. We achieve this without additional parameters. Comply yields sparse contextual representations of sentences that can be interpreted explicitly from the neuron weights.
2502.01707
CLIP-DQA: Blindly Evaluating Dehazed Images from Global and Local Perspectives Using CLIP
cs.CV cs.AI
Blind dehazed image quality assessment (BDQA), which aims to accurately predict the visual quality of dehazed images without any reference information, is essential for the evaluation, comparison, and optimization of image dehazing algorithms. Existing learning-based BDQA methods have achieved remarkable success, while the small scale of DQA datasets limits their performance. To address this issue, in this paper, we propose to adapt Contrastive Language-Image Pre-Training (CLIP), pre-trained on large-scale image-text pairs, to the BDQA task. Specifically, inspired by the fact that the human visual system understands images based on hierarchical features, we take global and local information of the dehazed image as the input of CLIP. To accurately map the input hierarchical information of dehazed images into the quality score, we tune both the vision branch and language branch of CLIP with prompt learning. Experimental results on two authentic DQA datasets demonstrate that our proposed approach, named CLIP-DQA, achieves more accurate quality predictions over existing BDQA methods. The code is available at https://github.com/JunFu1995/CLIP-DQA.
2502.01708
Aspects of Artificial Intelligence: Transforming Machine Learning Systems Naturally
cs.LG cs.AI cs.DB cs.DM
In this paper, we study the machine learning elements which we are interested in together as a machine learning system, consisting of a collection of machine learning elements and a collection of relations between the elements. The relations we concern are algebraic operations, binary relations, and binary relations with composition that can be reasoned categorically. A machine learning system transformation between two systems is a map between the systems, which preserves the relations we concern. The system transformations given by quotient or clustering, representable functor, and Yoneda embedding are highlighted and discussed by machine learning examples. An adjunction between machine learning systems, a special machine learning system transformation loop, provides the optimal way of solving problems. Machine learning system transformations are linked and compared by their maps at 2-cell, natural transformations. New insights and structures can be obtained from universal properties and algebraic structures given by monads, which are generated from adjunctions.
2502.01709
Adapter-Based Multi-Agent AVSR Extension for Pre-Trained ASR Models
cs.SD cs.LG eess.AS
We present an approach to Audio-Visual Speech Recognition that builds on a pre-trained Whisper model. To infuse visual information into this audio-only model, we extend it with an AV fusion module and LoRa adapters, one of the most up-to-date adapter approaches. One advantage of adapter-based approaches, is that only a relatively small number of parameters are trained, while the basic model remains unchanged. Common AVSR approaches train single models to handle several noise categories and noise levels simultaneously. Taking advantage of the lightweight nature of adapter approaches, we train noise-scenario-specific adapter-sets, each covering individual noise-categories or a specific noise-level range. The most suitable adapter-set is selected by previously classifying the noise-scenario. This enables our models to achieve an optimum coverage across different noise-categories and noise-levels, while training only a minimum number of parameters. Compared to a full fine-tuning approach with SOTA performance our models achieve almost comparable results over the majority of the tested noise-categories and noise-levels, with up to 88.5% less trainable parameters. Our approach can be extended by further noise-specific adapter-sets to cover additional noise scenarios. It is also possible to utilize the underlying powerful ASR model when no visual information is available, as it remains unchanged.
2502.01710
A Multi-Scale Feature Fusion Framework Integrating Frequency Domain and Cross-View Attention for Dual-View X-ray Security Inspections
cs.CV
With the rapid development of modern transportation systems and the exponential growth of logistics volumes, intelligent X-ray-based security inspection systems play a crucial role in public safety. Although single-view X-ray equipment is widely deployed, it struggles to accurately identify contraband in complex stacking scenarios due to strong viewpoint dependency and inadequate feature representation. To address this, we propose an innovative multi-scale interactive feature fusion framework tailored for dual-view X-ray security inspection image classification. The framework comprises three core modules: the Frequency Domain Interaction Module (FDIM) enhances frequency-domain features through Fourier transform; the Multi-Scale Cross-View Feature Enhancement (MSCFE) leverages cross-view attention mechanisms to strengthen feature interactions; and the Convolutional Attention Fusion Module (CAFM) efficiently fuses features by integrating channel attention with depthwise-separable convolutions. Experimental results demonstrate that our method outperforms existing state-of-the-art approaches across multiple backbone architectures, particularly excelling in complex scenarios with occlusions and object stacking.
2502.01711
Expected Return Symmetries
cs.MA
Symmetry is an important inductive bias that can improve model robustness and generalization across many deep learning domains. In multi-agent settings, a priori known symmetries have been shown to address a fundamental coordination failure mode known as mutually incompatible symmetry breaking; e.g. in a game where two independent agents can choose to move "left'' or "right'', and where a reward of +1 or -1 is received when the agents choose the same action or different actions, respectively. However, the efficient and automatic discovery of environment symmetries, in particular for decentralized partially observable Markov decision processes, remains an open problem. Furthermore, environmental symmetry breaking constitutes only one type of coordination failure, which motivates the search for a more accessible and broader symmetry class. In this paper, we introduce such a broader group of previously unexplored symmetries, which we call expected return symmetries, which contains environment symmetries as a subgroup. We show that agents trained to be compatible under the group of expected return symmetries achieve better zero-shot coordination results than those using environment symmetries. As an additional benefit, our method makes minimal a priori assumptions about the structure of their environment and does not require access to ground truth symmetries.
2502.01713
Auditing a Dutch Public Sector Risk Profiling Algorithm Using an Unsupervised Bias Detection Tool
cs.CY cs.LG
Algorithms are increasingly used to automate or aid human decisions, yet recent research shows that these algorithms may exhibit bias across legally protected demographic groups. However, data on these groups may be unavailable to organizations or external auditors due to privacy legislation. This paper studies bias detection using an unsupervised clustering tool when data on demographic groups are unavailable. We collaborate with the Dutch Executive Agency for Education to audit an algorithm that was used to assign risk scores to college students at the national level in the Netherlands between 2012-2023. Our audit covers more than 250,000 students from the whole country. The unsupervised clustering tool highlights known disparities between students with a non-European migration background and Dutch origin. Our contributions are three-fold: (1) we assess bias in a real-world, large-scale and high-stakes decision-making process by a governmental organization; (2) we use simulation studies to highlight potential pitfalls of using the unsupervised clustering tool to detect true bias when demographic group data are unavailable and provide recommendations for valid inferences; (3) we provide the unsupervised clustering tool in an open-source library. Our work serves as a starting point for a deliberative assessment by human experts to evaluate potential discrimination in algorithmic-supported decision-making processes.
2502.01714
Position: Towards a Responsible LLM-empowered Multi-Agent Systems
cs.MA cs.AI
The rise of Agent AI and Large Language Model-powered Multi-Agent Systems (LLM-MAS) has underscored the need for responsible and dependable system operation. Tools like LangChain and Retrieval-Augmented Generation have expanded LLM capabilities, enabling deeper integration into MAS through enhanced knowledge retrieval and reasoning. However, these advancements introduce critical challenges: LLM agents exhibit inherent unpredictability, and uncertainties in their outputs can compound across interactions, threatening system stability. To address these risks, a human-centered design approach with active dynamic moderation is essential. Such an approach enhances traditional passive oversight by facilitating coherent inter-agent communication and effective system governance, allowing MAS to achieve desired outcomes more efficiently.
2502.01715
Process-Supervised Reinforcement Learning for Code Generation
cs.SE cs.AI
Existing reinforcement learning strategies based on outcome supervision have proven effective in enhancing the performance of large language models(LLMs) for code generation. While reinforcement learning based on process supervision has shown great promise in handling multi-step reasoning tasks, its effectiveness in code generation remains largely underexplored and underjustified. The primary obstacle stems from the resource-intensive nature of constructing high-quality process-supervised data, which demands substantial human expertise and computational resources. In response to this challenge, we propose a "statement mutation/refactoring-compile and execution verification" strategy: mutating and refactoring code line-by-line through a teacher model, and utilizing compiler execution results to automatically label each line, resulting in line-by-line process-supervised data, which is pivotal for training a process-supervised reward model. The trained reward model is then integrated into the PRLCoder framework, followed by experimental validation on several benchmarks. Experimental results demonstrate that process-supervised reinforcement learning significantly surpasses methods relying solely on outcome supervision. Notably, in tackling complex code generation tasks, process-supervised reinforcement learning shows a clear advantage, ensuring both the integrity of the code generation process and the correctness of the generation results.
2502.01717
Choose Your Model Size: Any Compression by a Single Gradient Descent
cs.LG
The adoption of Foundation Models in resource-constrained environments remains challenging due to their large size and inference costs. A promising way to overcome these limitations is post-training compression, which aims to balance reduced model size against performance degradation. This work presents Any Compression via Iterative Pruning (ACIP), a novel algorithmic approach to determine a compression-performance trade-off from a single stochastic gradient descent run. To ensure parameter efficiency, we use an SVD-reparametrization of linear layers and iteratively prune their singular values with a sparsity-inducing penalty. The resulting pruning order gives rise to a global parameter ranking that allows us to materialize models of any target size. Importantly, the compressed models exhibit strong predictive downstream performance without the need for costly fine-tuning. We evaluate ACIP on a large selection of open-weight LLMs and tasks, and demonstrate state-of-the-art results compared to existing factorisation-based compression methods. We also show that ACIP seamlessly complements common quantization-based compression techniques.
2502.01718
ACECODER: Acing Coder RL via Automated Test-Case Synthesis
cs.SE cs.AI cs.CL
Most progress in recent coder models has been driven by supervised fine-tuning (SFT), while the potential of reinforcement learning (RL) remains largely unexplored, primarily due to the lack of reliable reward data/model in the code domain. In this paper, we address this challenge by leveraging automated large-scale test-case synthesis to enhance code model training. Specifically, we design a pipeline that generates extensive (question, test-cases) pairs from existing code data. Using these test cases, we construct preference pairs based on pass rates over sampled programs to train reward models with Bradley-Terry loss. It shows an average of 10-point improvement for Llama-3.1-8B-Ins and 5-point improvement for Qwen2.5-Coder-7B-Ins through best-of-32 sampling, making the 7B model on par with 236B DeepSeek-V2.5. Furthermore, we conduct reinforcement learning with both reward models and test-case pass rewards, leading to consistent improvements across HumanEval, MBPP, BigCodeBench, and LiveCodeBench (V4). Notably, we follow the R1-style training to start from Qwen2.5-Coder-base directly and show that our RL training can improve model on HumanEval-plus by over 25\% and MBPP-plus by 6\% for merely 80 optimization steps. We believe our results highlight the huge potential of reinforcement learning in coder models.
2502.01719
MJ-VIDEO: Fine-Grained Benchmarking and Rewarding Video Preferences in Video Generation
cs.CV
Recent advancements in video generation have significantly improved the ability to synthesize videos from text instructions. However, existing models still struggle with key challenges such as instruction misalignment, content hallucination, safety concerns, and bias. Addressing these limitations, we introduce MJ-BENCH-VIDEO, a large-scale video preference benchmark designed to evaluate video generation across five critical aspects: Alignment, Safety, Fineness, Coherence & Consistency, and Bias & Fairness. This benchmark incorporates 28 fine-grained criteria to provide a comprehensive evaluation of video preference. Building upon this dataset, we propose MJ-VIDEO, a Mixture-of-Experts (MoE)-based video reward model designed to deliver fine-grained reward. MJ-VIDEO can dynamically select relevant experts to accurately judge the preference based on the input text-video pair. This architecture enables more precise and adaptable preference judgments. Through extensive benchmarking on MJ-BENCH-VIDEO, we analyze the limitations of existing video reward models and demonstrate the superior performance of MJ-VIDEO in video preference assessment, achieving 17.58% and 15.87% improvements in overall and fine-grained preference judgments, respectively. Additionally, introducing MJ-VIDEO for preference tuning in video generation enhances the alignment performance. All our code, data, and models are available at https://aiming-lab.github.io/MJ-VIDEO.github.io/.
2502.01720
Generating Multi-Image Synthetic Data for Text-to-Image Customization
cs.CV cs.GR cs.LG
Customization of text-to-image models enables users to insert custom concepts and generate the concepts in unseen settings. Existing methods either rely on costly test-time optimization or train encoders on single-image training datasets without multi-image supervision, leading to worse image quality. We propose a simple approach that addresses both limitations. We first leverage existing text-to-image models and 3D datasets to create a high-quality Synthetic Customization Dataset (SynCD) consisting of multiple images of the same object in different lighting, backgrounds, and poses. We then propose a new encoder architecture based on shared attention mechanisms that better incorporate fine-grained visual details from input images. Finally, we propose a new inference technique that mitigates overexposure issues during inference by normalizing the text and image guidance vectors. Through extensive experiments, we show that our model, trained on the synthetic dataset with the proposed encoder and inference algorithm, outperforms existing tuning-free methods on standard customization benchmarks.
2502.01739
Grokking vs. Learning: Same Features, Different Encodings
cs.LG cond-mat.dis-nn cs.AI
Grokking typically achieves similar loss to ordinary, "steady", learning. We ask whether these different learning paths - grokking versus ordinary training - lead to fundamental differences in the learned models. To do so we compare the features, compressibility, and learning dynamics of models trained via each path in two tasks. We find that grokked and steadily trained models learn the same features, but there can be large differences in the efficiency with which these features are encoded. In particular, we find a novel "compressive regime" of steady training in which there emerges a linear trade-off between model loss and compressibility, and which is absent in grokking. In this regime, we can achieve compression factors 25x times the base model, and 5x times the compression achieved in grokking. We then track how model features and compressibility develop through training. We show that model development in grokking is task-dependent, and that peak compressibility is achieved immediately after the grokking plateau. Finally, novel information-geometric measures are introduced which demonstrate that models undergoing grokking follow a straight path in information space.
2502.01754
Evaluation of Large Language Models via Coupled Token Generation
cs.CL cs.AI cs.LG
State of the art large language models rely on randomization to respond to a prompt. As an immediate consequence, a model may respond differently to the same prompt if asked multiple times. In this work, we argue that the evaluation and ranking of large language models should control for the randomization underpinning their functioning. Our starting point is the development of a causal model for coupled autoregressive generation, which allows different large language models to sample responses with the same source of randomness. Building upon our causal model, we first show that, on evaluations based on benchmark datasets, coupled autoregressive generation leads to the same conclusions as vanilla autoregressive generation but using provably fewer samples. However, we further show that, on evaluations based on (human) pairwise comparisons, coupled and vanilla autoregressive generation can surprisingly lead to different rankings when comparing more than two models, even with an infinite amount of samples. This suggests that the apparent advantage of a model over others in existing evaluation protocols may not be genuine but rather confounded by the randomness inherent to the generation process. To illustrate and complement our theoretical results, we conduct experiments with several large language models from the Llama family. We find that, across multiple knowledge areas from the popular MMLU benchmark dataset, coupled autoregressive generation requires up to 40% fewer samples to reach the same conclusions as vanilla autoregressive generation. Further, using data from the LMSYS Chatbot Arena platform, we find that the win-rates derived from pairwise comparisons by a strong large language model to prompts differ under coupled and vanilla autoregressive generation.
2502.01755
Robust Federated Finetuning of LLMs via Alternating Optimization of LoRA
cs.LG cs.AI
Parameter-Efficient Fine-Tuning (PEFT) methods like Low-Rank Adaptation (LoRA) optimize federated training by reducing computational and communication costs. We propose RoLoRA, a federated framework using alternating optimization to fine-tune LoRA adapters. Our approach emphasizes the importance of learning up and down projection matrices to enhance expressiveness and robustness. We use both theoretical analysis and extensive experiments to demonstrate the advantages of RoLoRA over prior approaches that either generate imperfect model updates or limit expressiveness of the model. We present theoretical analysis on a simplified linear model to demonstrate the importance of learning both down-projection and up-projection matrices in LoRA. We provide extensive experimental evaluations on a toy neural network on MNIST as well as large language models including RoBERTa-Large, Llama-2-7B on diverse tasks to demonstrate the advantages of RoLoRA over other methods.
2502.01763
On The Concurrence of Layer-wise Preconditioning Methods and Provable Feature Learning
cs.LG math.OC stat.ML
Layer-wise preconditioning methods are a family of memory-efficient optimization algorithms that introduce preconditioners per axis of each layer's weight tensors. These methods have seen a recent resurgence, demonstrating impressive performance relative to entry-wise ("diagonal") preconditioning methods such as Adam(W) on a wide range of neural network optimization tasks. Complementary to their practical performance, we demonstrate that layer-wise preconditioning methods are provably necessary from a statistical perspective. To showcase this, we consider two prototypical models, linear representation learning and single-index learning, which are widely used to study how typical algorithms efficiently learn useful features to enable generalization. In these problems, we show SGD is a suboptimal feature learner when extending beyond ideal isotropic inputs $\mathbf{x} \sim \mathsf{N}(\mathbf{0}, \mathbf{I})$ and well-conditioned settings typically assumed in prior work. We demonstrate theoretically and numerically that this suboptimality is fundamental, and that layer-wise preconditioning emerges naturally as the solution. We further show that standard tools like Adam preconditioning and batch-norm only mildly mitigate these issues, supporting the unique benefits of layer-wise preconditioning.
2502.01770
Hamming Attention Distillation: Binarizing Keys and Queries for Efficient Long-Context Transformers
cs.LG cs.AI eess.IV
Pre-trained transformer models with extended context windows are notoriously expensive to run at scale, often limiting real-world deployment due to their high computational and memory requirements. In this paper, we introduce Hamming Attention Distillation (HAD), a novel framework that binarizes keys and queries in the attention mechanism to achieve significant efficiency gains. By converting keys and queries into {-1, +1} vectors and replacing dot-product operations with efficient Hamming distance computations, our method drastically reduces computational overhead. Additionally, we incorporate attention matrix sparsification to prune low-impact activations, which further reduces the cost of processing long-context sequences. \par Despite these aggressive compression strategies, our distilled approach preserves a high degree of representational power, leading to substantially improved accuracy compared to prior transformer binarization methods. We evaluate HAD on a range of tasks and models, including the GLUE benchmark, ImageNet, and QuALITY, demonstrating state-of-the-art performance among binarized Transformers while drastically reducing the computational costs of long-context inference. \par We implement HAD in custom hardware simulations, demonstrating superior performance characteristics compared to a custom hardware implementation of standard attention. HAD achieves just $\mathbf{1.78}\%$ performance losses on GLUE compared to $9.08\%$ in state-of-the-art binarization work, and $\mathbf{2.5}\%$ performance losses on ImageNet compared to $12.14\%$, all while targeting custom hardware with a $\mathbf{79}\%$ area reduction and $\mathbf{87}\%$ power reduction compared to its standard attention counterpart.
2502.01772
On Bob Dylan: A Computational Perspective
cs.CL cs.AI cs.IR cs.SI
Cass Sunstein's essay 'On Bob Dylan' describes Dylan's 'dishabituating' style -- a constant refusal to conform to expectation and a penchant for reinventing his musical and lyrical identity. In this paper, I extend Sunstein's observations through a large-scale computational analysis of Dylan's lyrics from 1962 to 2012. Using o3-mini-high (a large language model), I extract concept-to-concept relationships from the lyrics and construct directed knowledge graphs that capture Dylan's thematic structure. I then quantify shifts in sentiment, metaphorical expression, thematic diversity, and network complexity over time. The results indicate that Dylan's lyrics increasingly rely on metaphor, display an evolving sentiment profile, and exhibit heightened dishabituation -- measured here as a growing variance in the network centrality of key concepts. I also find that references to movement, protest, and mythic imagery fluctuate in ways that align with well-known phases of Dylan's career, reflecting the dynamic and unpredictable quality of his art. These findings not only deepen our empirical understanding of Sunstein's thesis but also introduce a novel computational method for analyzing an artist's evolution-offering broader applicability to the study of cultural and creative change.
2502.01773
Coarse-to-Fine 3D Keyframe Transporter
cs.RO cs.CV
Recent advances in Keyframe Imitation Learning (IL) have enabled learning-based agents to solve a diverse range of manipulation tasks. However, most approaches ignore the rich symmetries in the problem setting and, as a consequence, are sample-inefficient. This work identifies and utilizes the bi-equivariant symmetry within Keyframe IL to design a policy that generalizes to transformations of both the workspace and the objects grasped by the gripper. We make two main contributions: First, we analyze the bi-equivariance properties of the keyframe action scheme and propose a Keyframe Transporter derived from the Transporter Networks, which evaluates actions using cross-correlation between the features of the grasped object and the features of the scene. Second, we propose a computationally efficient coarse-to-fine SE(3) action evaluation scheme for reasoning the intertwined translation and rotation action. The resulting method outperforms strong Keyframe IL baselines by an average of >10% on a wide range of simulation tasks, and by an average of 55% in 4 physical experiments.
2502.01774
Grokking Explained: A Statistical Phenomenon
cs.LG cs.AI
Grokking, or delayed generalization, is an intriguing learning phenomenon where test set loss decreases sharply only after a model's training set loss has converged. This challenges conventional understanding of the training dynamics in deep learning networks. In this paper, we formalize and investigate grokking, highlighting that a key factor in its emergence is a distribution shift between training and test data. We introduce two synthetic datasets specifically designed to analyze grokking. One dataset examines the impact of limited sampling, and the other investigates transfer learning's role in grokking. By inducing distribution shifts through controlled imbalanced sampling of sub-categories, we systematically reproduce the phenomenon, demonstrating that while small-sampling is strongly associated with grokking, it is not its cause. Instead, small-sampling serves as a convenient mechanism for achieving the necessary distribution shift. We also show that when classes form an equivariant map, grokking can be explained by the model's ability to learn from similar classes or sub-categories. Unlike earlier work suggesting that grokking primarily arises from high regularization and sparse data, we demonstrate that it can also occur with dense data and minimal hyper-parameter tuning. Our findings deepen the understanding of grokking and pave the way for developing better stopping criteria in future training processes.
2502.01776
Sparse VideoGen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity
cs.CV cs.LG
Diffusion Transformers (DiTs) dominate video generation but their high computational cost severely limits real-world applicability, usually requiring tens of minutes to generate a few seconds of video even on high-performance GPUs. This inefficiency primarily arises from the quadratic computational complexity of 3D Full Attention with respect to the context length. In this paper, we propose a training-free framework termed Sparse VideoGen (SVG) that leverages the inherent sparsity in 3D Full Attention to boost inference efficiency. We reveal that the attention heads can be dynamically classified into two groups depending on distinct sparse patterns: (1) Spatial Head, where only spatially-related tokens within each frame dominate the attention output, and (2) Temporal Head, where only temporally-related tokens across different frames dominate. Based on this insight, SVG proposes an online profiling strategy to capture the dynamic sparse patterns and predicts the type of attention head. Combined with a novel hardware-efficient tensor layout transformation and customized kernel implementations, SVG achieves up to 2.28x and 2.33x end-to-end speedup on CogVideoX-v1.5 and HunyuanVideo, respectively, while preserving generation quality.
2502.01777
CTC-DRO: Robust Optimization for Reducing Language Disparities in Speech Recognition
cs.LG cs.CL eess.AS
Modern deep learning models often achieve high overall performance, but consistently fail on specific subgroups. Group distributionally robust optimization (group DRO) addresses this problem by minimizing the worst-group loss, but it fails when group losses misrepresent performance differences between groups. This is common in domains like speech, where the widely used connectionist temporal classification (CTC) loss scales with input length and varies with linguistic and acoustic properties, leading to spurious differences between group losses. We present CTC-DRO, which addresses the shortcomings of the group DRO objective by smoothing the group weight update to prevent overemphasis on consistently high-loss groups, while using input length-matched batching to mitigate CTC's scaling issues. We evaluate CTC-DRO on the task of multilingual automatic speech recognition (ASR) across five language sets from the ML-SUPERB 2.0 benchmark. CTC-DRO consistently outperforms group DRO and CTC-based baseline models, reducing the worst-language error by up to 65.9% and the average error by up to 47.7%. CTC-DRO can be applied to ASR with minimal computational costs, and offers the potential for reducing group disparities in other domains with similar challenges.
2502.01778
GNN-DT: Graph Neural Network Enhanced Decision Transformer for Efficient Optimization in Dynamic Environments
cs.LG cs.SY eess.SY
Reinforcement Learning (RL) methods used for solving real-world optimization problems often involve dynamic state-action spaces, larger scale, and sparse rewards, leading to significant challenges in convergence, scalability, and efficient exploration of the solution space. This study introduces GNN-DT, a novel Decision Transformer (DT) architecture that integrates Graph Neural Network (GNN) embedders with a novel residual connection between input and output tokens crucial for handling dynamic environments. By learning from previously collected trajectories, GNN-DT reduces dependence on accurate simulators and tackles the sparse rewards limitations of online RL algorithms. We evaluate GNN-DT on the complex electric vehicle (EV) charging optimization problem and prove that its performance is superior and requires significantly fewer training trajectories, thus improving sample efficiency compared to existing DT baselines. Furthermore, GNN-DT exhibits robust generalization to unseen environments and larger action spaces, addressing a critical gap in prior DT-based approaches
2502.01780
Graph Canonical Correlation Analysis
stat.ML cs.LG
Canonical correlation analysis (CCA) is a widely used technique for estimating associations between two sets of multi-dimensional variables. Recent advancements in CCA methods have expanded their application to decipher the interactions of multiomics datasets, imaging-omics datasets, and more. However, conventional CCA methods are limited in their ability to incorporate structured patterns in the cross-correlation matrix, potentially leading to suboptimal estimations. To address this limitation, we propose the graph Canonical Correlation Analysis (gCCA) approach, which calculates canonical correlations based on the graph structure of the cross-correlation matrix between the two sets of variables. We develop computationally efficient algorithms for gCCA, and provide theoretical results for finite sample analysis of best subset selection and canonical correlation estimation by introducing concentration inequalities and stopping time rule based on martingale theories. Extensive simulations demonstrate that gCCA outperforms competing CCA methods. Additionally, we apply gCCA to a multiomics dataset of DNA methylation and RNA-seq transcriptomics, identifying both positively and negatively regulated gene expression pathways by DNA methylation pathways.
2502.01784
VILP: Imitation Learning with Latent Video Planning
cs.RO cs.CV
In the era of generative AI, integrating video generation models into robotics opens new possibilities for the general-purpose robot agent. This paper introduces imitation learning with latent video planning (VILP). We propose a latent video diffusion model to generate predictive robot videos that adhere to temporal consistency to a good degree. Our method is able to generate highly time-aligned videos from multiple views, which is crucial for robot policy learning. Our video generation model is highly time-efficient. For example, it can generate videos from two distinct perspectives, each consisting of six frames with a resolution of 96x160 pixels, at a rate of 5 Hz. In the experiments, we demonstrate that VILP outperforms the existing video generation robot policy across several metrics: training costs, inference speed, temporal consistency of generated videos, and the performance of the policy. We also compared our method with other imitation learning methods. Our findings indicate that VILP can rely less on extensive high-quality task-specific robot action data while still maintaining robust performance. In addition, VILP possesses robust capabilities in representing multi-modal action distributions. Our paper provides a practical example of how to effectively integrate video generation models into robot policies, potentially offering insights for related fields and directions. For more details, please refer to our open-source repository https://github.com/ZhengtongXu/VILP.
2502.01785
AquaticCLIP: A Vision-Language Foundation Model for Underwater Scene Analysis
cs.CV cs.AI
The preservation of aquatic biodiversity is critical in mitigating the effects of climate change. Aquatic scene understanding plays a pivotal role in aiding marine scientists in their decision-making processes. In this paper, we introduce AquaticCLIP, a novel contrastive language-image pre-training model tailored for aquatic scene understanding. AquaticCLIP presents a new unsupervised learning framework that aligns images and texts in aquatic environments, enabling tasks such as segmentation, classification, detection, and object counting. By leveraging our large-scale underwater image-text paired dataset without the need for ground-truth annotations, our model enriches existing vision-language models in the aquatic domain. For this purpose, we construct a 2 million underwater image-text paired dataset using heterogeneous resources, including YouTube, Netflix, NatGeo, etc. To fine-tune AquaticCLIP, we propose a prompt-guided vision encoder that progressively aggregates patch features via learnable prompts, while a vision-guided mechanism enhances the language encoder by incorporating visual context. The model is optimized through a contrastive pretraining loss to align visual and textual modalities. AquaticCLIP achieves notable performance improvements in zero-shot settings across multiple underwater computer vision tasks, outperforming existing methods in both robustness and interpretability. Our model sets a new benchmark for vision-language applications in underwater environments. The code and dataset for AquaticCLIP are publicly available on GitHub at xxx.
2502.01787
The Effects of Enterprise Social Media on Communication Networks
cs.CY cs.SI
Enterprise social media platforms (ESMPs) are web-based platforms with standard social media functionality, e.g., communicating with others, posting links and files, liking content, etc., yet all users are part of the same company. The first contribution of this work is the use of a difference-in-differences analysis of $99$ companies to measure the causal impact of ESMPs on companies' communication networks across the full spectrum of communication technologies used within companies: email, instant messaging, and ESMPs. Adoption caused companies' communication networks to grow denser and more well-connected by adding new, novel ties that often, but not exclusively, involve communication from one to many employees. Importantly, some new ties also bridge otherwise separate parts of the corporate communication network. The second contribution of this work, utilizing data on Microsoft's own communication network, is understanding how these communication technologies connect people across the corporate hierarchy. Compared to email and instant messaging, ESMPs excel at connecting nodes distant in the corporate hierarchy both vertically (between leaders and employees) and horizontally (between employees in similar roles but different sectors). Also, influence in ESMPs is more `democratic' than elsewhere, with high-influence nodes well-distributed across the corporate hierarchy. Overall, our results suggest that ESMPs boost information flow within companies and increase employees' attention to what is happening outside their immediate working group, above and beyond email and instant messaging.
2502.01789
An Agentic AI Workflow for Detecting Cognitive Concerns in Real-world Data
cs.AI cs.MA
Early identification of cognitive concerns is critical but often hindered by subtle symptom presentation. This study developed and validated a fully automated, multi-agent AI workflow using LLaMA 3 8B to identify cognitive concerns in 3,338 clinical notes from Mass General Brigham. The agentic workflow, leveraging task-specific agents that dynamically collaborate to extract meaningful insights from clinical notes, was compared to an expert-driven benchmark. Both workflows achieved high classification performance, with F1-scores of 0.90 and 0.91, respectively. The agentic workflow demonstrated improved specificity (1.00) and achieved prompt refinement in fewer iterations. Although both workflows showed reduced performance on validation data, the agentic workflow maintained perfect specificity. These findings highlight the potential of fully automated multi-agent AI workflows to achieve expert-level accuracy with greater efficiency, offering a scalable and cost-effective solution for detecting cognitive concerns in clinical settings.
2502.01792
Policy Design for Two-sided Platforms with Participation Dynamics
cs.GT cs.IR cs.LG cs.SY eess.SY
In two-sided platforms (e.g., video streaming or e-commerce), viewers and providers engage in interactive dynamics, where an increased provider population results in higher viewer utility and the increase of viewer population results in higher provider utility. Despite the importance of such "population effects" on long-term platform health, recommendation policies do not generally take the participation dynamics into account. This paper thus studies the dynamics and policy design on two-sided platforms under the population effects for the first time. Our control- and game-theoretic findings warn against the use of myopic-greedy policy and shed light on the importance of provider-side considerations (i.e., effectively distributing exposure among provider groups) to improve social welfare via population growth. We also present a simple algorithm to optimize long-term objectives by considering the population effects, and demonstrate its effectiveness in synthetic and real-data experiments.
2502.01800
Flow-based Domain Randomization for Learning and Sequencing Robotic Skills
cs.RO cs.AI cs.LG
Domain randomization in reinforcement learning is an established technique for increasing the robustness of control policies trained in simulation. By randomizing environment properties during training, the learned policy can become robust to uncertainties along the randomized dimensions. While the environment distribution is typically specified by hand, in this paper we investigate automatically discovering a sampling distribution via entropy-regularized reward maximization of a normalizing-flow-based neural sampling distribution. We show that this architecture is more flexible and provides greater robustness than existing approaches that learn simpler, parameterized sampling distributions, as demonstrated in six simulated and one real-world robotics domain. Lastly, we explore how these learned sampling distributions, combined with a privileged value function, can be used for out-of-distribution detection in an uncertainty-aware multi-step manipulation planner.
2502.01803
Discovering Chunks in Neural Embeddings for Interpretability
cs.LG cs.AI
Understanding neural networks is challenging due to their high-dimensional, interacting components. Inspired by human cognition, which processes complex sensory data by chunking it into recurring entities, we propose leveraging this principle to interpret artificial neural population activities. Biological and artificial intelligence share the challenge of learning from structured, naturalistic data, and we hypothesize that the cognitive mechanism of chunking can provide insights into artificial systems. We first demonstrate this concept in recurrent neural networks (RNNs) trained on artificial sequences with imposed regularities, observing that their hidden states reflect these patterns, which can be extracted as a dictionary of chunks that influence network responses. Extending this to large language models (LLMs) like LLaMA, we identify similar recurring embedding states corresponding to concepts in the input, with perturbations to these states activating or inhibiting the associated concepts. By exploring methods to extract dictionaries of identifiable chunks across neural embeddings of varying complexity, our findings introduce a new framework for interpreting neural networks, framing their population activity as structured reflections of the data they process.
2502.01804
Soup-of-Experts: Pretraining Specialist Models via Parameters Averaging
cs.LG cs.CL
Machine learning models are routinely trained on a mixture of different data domains. Different domain weights yield very different downstream performances. We propose the Soup-of-Experts, a novel architecture that can instantiate a model at test time for any domain weights with minimal computational cost and without re-training the model. Our architecture consists of a bank of expert parameters, which are linearly combined to instantiate one model. We learn the linear combination coefficients as a function of the input domain weights. To train this architecture, we sample random domain weights, instantiate the corresponding model, and backprop through one batch of data sampled with these domain weights. We demonstrate how our approach obtains small specialized models on several language modeling tasks quickly. Soup-of-Experts are particularly appealing when one needs to ship many different specialist models quickly under a model size constraint.
2502.01806
Toward Neurosymbolic Program Comprehension
cs.SE cs.AI
Recent advancements in Large Language Models (LLMs) have paved the way for Large Code Models (LCMs), enabling automation in complex software engineering tasks, such as code generation, software testing, and program comprehension, among others. Tools like GitHub Copilot and ChatGPT have shown substantial benefits in supporting developers across various practices. However, the ambition to scale these models to trillion-parameter sizes, exemplified by GPT-4, poses significant challenges that limit the usage of Artificial Intelligence (AI)-based systems powered by large Deep Learning (DL) models. These include rising computational demands for training and deployment and issues related to trustworthiness, bias, and interpretability. Such factors can make managing these models impractical for many organizations, while their "black-box'' nature undermines key aspects, including transparency and accountability. In this paper, we question the prevailing assumption that increasing model parameters is always the optimal path forward, provided there is sufficient new data to learn additional patterns. In particular, we advocate for a Neurosymbolic research direction that combines the strengths of existing DL techniques (e.g., LLMs) with traditional symbolic methods--renowned for their reliability, speed, and determinism. To this end, we outline the core features and present preliminary results for our envisioned approach, aimed at establishing the first Neurosymbolic Program Comprehension (NsPC) framework to aid in identifying defective code components.
2502.01809
Self-supervised Subgraph Neural Network With Deep Reinforcement Walk Exploration
cs.LG
Graph data, with its structurally variable nature, represents complex real-world phenomena like chemical compounds, protein structures, and social networks. Traditional Graph Neural Networks (GNNs) primarily utilize the message-passing mechanism, but their expressive power is limited and their prediction lacks explainability. To address these limitations, researchers have focused on graph substructures. Subgraph neural networks (SGNNs) and GNN explainers have emerged as potential solutions, but each has its limitations. SGNNs computes graph representations based on the bags of subgraphs to enhance the expressive power. However, they often rely on predefined algorithm-based sampling strategies, which is inefficient. GNN explainers adopt data-driven approaches to generate important subgraphs to provide explanation. Nevertheless, their explanation is difficult to be translated into practical improvements on GNNs. To overcome these issues, we propose a novel self-supervised framework that integrates SGNNs with the generation approach of GNN explainers, named the Reinforcement Walk Exploration SGNN (RWE-SGNN). Our approach features a sampling model trained in an explainer fashion, optimizing subgraphs to enhance model performance. To achieve a data-driven sampling approach, unlike traditional subgraph generation approaches, we propose a novel walk exploration process, which efficiently extracts important substructures, simplifying the embedding process and avoiding isomorphism problems. Moreover, we prove that our proposed walk exploration process has equivalent generation capability to the traditional subgraph generation process. Experimental results on various graph datasets validate the effectiveness of our proposed method, demonstrating significant improvements in performance and precision.
2502.01810
Estimating Network Models using Neural Networks
cs.SI econ.EM stat.CO stat.ML
Exponential random graph models (ERGMs) are very flexible for modeling network formation but pose difficult estimation challenges due to their intractable normalizing constant. Existing methods, such as MCMC-MLE, rely on sequential simulation at every optimization step. We propose a neural network approach that trains on a single, large set of parameter-simulation pairs to learn the mapping from parameters to average network statistics. Once trained, this map can be inverted, yielding a fast and parallelizable estimation method. The procedure also accommodates extra network statistics to mitigate model misspecification. Some simple illustrative examples show that the method performs well in practice.
2502.01812
SelfCheckAgent: Zero-Resource Hallucination Detection in Generative Large Language Models
cs.CL cs.LG
Detecting hallucinations in Large Language Models (LLMs) remains a critical challenge for their reliable deployment in real-world applications. To address this, we introduce SelfCheckAgent, a novel framework integrating three different agents: the Symbolic Agent, the Specialized Detection Agent, and the Contextual Consistency Agent. These agents provide a robust multi-dimensional approach to hallucination detection. Notable results include the Contextual Consistency Agent leveraging Llama 3.1 with Chain-of-Thought (CoT) to achieve outstanding performance on the WikiBio dataset, with NonFactual hallucination detection scoring 93.64%, Factual 70.26%, and Ranking 78.48% respectively. On the AIME dataset, GPT-4o with CoT excels in NonFactual detection with 94.89% but reveals trade-offs in Factual with 30.58% and Ranking with 30.68%, underscoring the complexity of hallucination detection in the complex mathematical domains. The framework also incorporates a triangulation strategy, which increases the strengths of the SelfCheckAgent, yielding significant improvements in real-world hallucination identification. The comparative analysis demonstrates SelfCheckAgent's applicability across diverse domains, positioning it as a crucial advancement for trustworthy LLMs. These findings highlight the potentiality of consistency-driven methodologies in detecting hallucinations in LLMs.
2502.01814
PolyhedronNet: Representation Learning for Polyhedra with Surface-attributed Graph
cs.CV cs.LG
Ubiquitous geometric objects can be precisely and efficiently represented as polyhedra. The transformation of a polyhedron into a vector, known as polyhedra representation learning, is crucial for manipulating these shapes with mathematical and statistical tools for tasks like classification, clustering, and generation. Recent years have witnessed significant strides in this domain, yet most efforts focus on the vertex sequence of a polyhedron, neglecting the complex surface modeling crucial in real-world polyhedral objects. This study proposes \textbf{PolyhedronNet}, a general framework tailored for learning representations of 3D polyhedral objects. We propose the concept of the surface-attributed graph to seamlessly model the vertices, edges, faces, and their geometric interrelationships within a polyhedron. To effectively learn the representation of the entire surface-attributed graph, we first propose to break it down into local rigid representations to effectively learn each local region's relative positions against the remaining regions without geometric information loss. Subsequently, we propose PolyhedronGNN to hierarchically aggregate the local rigid representation via intra-face and inter-face geometric message passing modules, to obtain a global representation that minimizes information loss while maintaining rotation and translation invariance. Our experimental evaluations on four distinct datasets, encompassing both classification and retrieval tasks, substantiate PolyhedronNet's efficacy in capturing comprehensive and informative representations of 3D polyhedral objects. Code and data are available at {https://github.com/dyu62/3D_polyhedron}.
2502.01816
Low Resource Video Super-resolution using Memory and Residual Deformable Convolutions
cs.CV cs.LG
Transformer-based video super-resolution (VSR) models have set new benchmarks in recent years, but their substantial computational demands make most of them unsuitable for deployment on resource-constrained devices. Achieving a balance between model complexity and output quality remains a formidable challenge in VSR. Although lightweight models have been introduced to address this issue, they often struggle to deliver state-of-the-art performance. We propose a novel lightweight, parameter-efficient deep residual deformable convolution network for VSR. Unlike prior methods, our model enhances feature utilization through residual connections and employs deformable convolution for precise frame alignment, addressing motion dynamics effectively. Furthermore, we introduce a single memory tensor to capture information accrued from the past frames and improve motion estimation across frames. This design enables an efficient balance between computational cost and reconstruction quality. With just 2.3 million parameters, our model achieves state-of-the-art SSIM of 0.9175 on the REDS4 dataset, surpassing existing lightweight and many heavy models in both accuracy and resource efficiency. Architectural insights from our model pave the way for real-time VSR on streaming data.
2502.01819
Score as Action: Fine-Tuning Diffusion Generative Models by Continuous-time Reinforcement Learning
cs.LG cs.AI math.OC
Reinforcement learning from human feedback (RLHF), which aligns a diffusion model with input prompt, has become a crucial step in building reliable generative AI models. Most works in this area use a discrete-time formulation, which is prone to induced errors, and often not applicable to models with higher-order/black-box solvers. The objective of this study is to develop a disciplined approach to fine-tune diffusion models using continuous-time RL, formulated as a stochastic control problem with a reward function that aligns the end result (terminal state) with input prompt. The key idea is to treat score matching as controls or actions, and thereby making connections to policy optimization and regularization in continuous-time RL. To carry out this idea, we lay out a new policy optimization framework for continuous-time RL, and illustrate its potential in enhancing the value networks design space via leveraging the structural property of diffusion models. We validate the advantages of our method by experiments in downstream tasks of fine-tuning large-scale Text2Image models of Stable Diffusion v1.5.
2502.01820
Physics-Informed Surrogates for Temperature Prediction of Multi-Tracks in Laser Powder Bed Fusion
cs.CE
Modeling plays a critical role in additive manufacturing (AM), enabling a deeper understanding of underlying processes. Parametric solutions for such models are of great importance, enabling the optimization of production processes and considerable cost reductions. However, the complexity of the problem and diversity of spatio-temporal scales involved in the process pose significant challenges for traditional numerical methods. Surrogate models offer a powerful alternative by accelerating simulations and facilitating real-time monitoring and control. The present study presents an operator learning approach that relies on the deep operator network (DeepONet) and physics-informed neural networks (PINN) to predict the three-dimensional temperature distribution during melting and consolidation in laser powder bed fusion (LPBF). Parametric solutions for both single-track and multi-track scenarios with respect to tool path are obtained. To address the challenges in obtaining parametric solutions for multi-track scenarios using DeepONet architecture, a sequential PINN approach is proposed to efficiently manage the increased training complexity inherent in those scenarios. The accuracy and consistency of the model are verified against finite-difference computations. The developed surrogate allows us to efficiently analyze the effect of scanning paths and laser parameters on the thermal history.
2502.01821
Agentic Bug Reproduction for Effective Automated Program Repair at Google
cs.SE cs.AI
Bug reports often lack sufficient detail for developers to reproduce and fix the underlying defects. Bug Reproduction Tests (BRTs), tests that fail when the bug is present and pass when it has been resolved, are crucial for debugging, but they are rarely included in bug reports, both in open-source and in industrial settings. Thus, automatically generating BRTs from bug reports has the potential to accelerate the debugging process and lower time to repair. This paper investigates automated BRT generation within an industry setting, specifically at Google, focusing on the challenges of a large-scale, proprietary codebase and considering real-world industry bugs extracted from Google's internal issue tracker. We adapt and evaluate a state-of-the-art BRT generation technique, LIBRO, and present our agent-based approach, BRT Agent, which makes use of a fine-tuned Large Language Model (LLM) for code editing. Our BRT Agent significantly outperforms LIBRO, achieving a 28% plausible BRT generation rate, compared to 10% by LIBRO, on 80 human-reported bugs from Google's internal issue tracker. We further investigate the practical value of generated BRTs by integrating them with an Automated Program Repair (APR) system at Google. Our results show that providing BRTs to the APR system results in 30% more bugs with plausible fixes. Additionally, we introduce Ensemble Pass Rate (EPR), a metric which leverages the generated BRTs to select the most promising fixes from all fixes generated by APR system. Our evaluation on EPR for Top-K and threshold-based fix selections demonstrates promising results and trade-offs. For example, EPR correctly selects a plausible fix from a pool of 20 candidates in 70% of cases, based on its top-1 ranking.
2502.01825
Assessing Data Augmentation-Induced Bias in Training and Testing of Machine Learning Models
cs.SE cs.AI
Data augmentation has become a standard practice in software engineering to address limited or imbalanced data sets, particularly in specialized domains like test classification and bug detection where data can be scarce. Although techniques such as SMOTE and mutation-based augmentation are widely used in software testing and debugging applications, a rigorous understanding of how augmented training data impacts model bias is lacking. It is especially critical to consider bias in scenarios where augmented data sets are used not just in training but also in testing models. Through a comprehensive case study of flaky test classification, we demonstrate how to test for bias and understand the impact that the inclusion of augmented samples in testing sets can have on model evaluation.
2502.01827
Relatively-Secure LLM-Based Steganography via Constrained Markov Decision Processes
cs.IT math.IT
Linguistic steganography aims to conceal information within natural language text without being detected. An effective steganography approach should encode the secret message into a minimal number of language tokens while preserving the natural appearance and fluidity of the stego-texts. We present a new framework to enhance the embedding efficiency of stego-texts generated by modifying the output of a large language model (LLM). The novelty of our approach is in abstracting the sequential steganographic embedding process as a Constrained Markov Decision Process (CMDP), which takes into consideration the long-term dependencies instead of merely the immediate effects. We constrain the solution space such that the discounted accumulative total variation divergence between the selected probability distribution and the original distribution given by the LLM is below a threshold. To find the optimal policy, we first show that the functional optimization problem can be simplified to a convex optimization problem with a finite number of variables. A closed-form solution for the optimal policy is then presented to this equivalent problem. It is remarkable that the optimal policy is deterministic and resembles water-filling in some cases. The solution suggests that usually adjusting the probability distribution for the state that has the least random transition probability should be prioritized, but the choice should be made by taking into account the transition probabilities at all states instead of only the current state.
2502.01828
From Foresight to Forethought: VLM-In-the-Loop Policy Steering via Latent Alignment
cs.RO cs.LG
While generative robot policies have demonstrated significant potential in learning complex, multimodal behaviors from demonstrations, they still exhibit diverse failures at deployment-time. Policy steering offers an elegant solution to reducing the chance of failure by using an external verifier to select from low-level actions proposed by an imperfect generative policy. Here, one might hope to use a Vision Language Model (VLM) as a verifier, leveraging its open-world reasoning capabilities. However, off-the-shelf VLMs struggle to understand the consequences of low-level robot actions as they are represented fundamentally differently than the text and images the VLM was trained on. In response, we propose FOREWARN, a novel framework to unlock the potential of VLMs as open-vocabulary verifiers for runtime policy steering. Our key idea is to decouple the VLM's burden of predicting action outcomes (foresight) from evaluation (forethought). For foresight, we leverage a latent world model to imagine future latent states given diverse low-level action plans. For forethought, we align the VLM with these predicted latent states to reason about the consequences of actions in its native representation--natural language--and effectively filter proposed plans. We validate our framework across diverse robotic manipulation tasks, demonstrating its ability to bridge representational gaps and provide robust, generalizable policy steering. Videos can be found on the project website: https://yilin-wu98.github.io/forewarn/.
2502.01830
Meta-neural Topology Optimization: Knowledge Infusion with Meta-learning
cs.CE physics.comp-ph
Engineers learn from every design they create, building intuition that helps them quickly identify promising solutions for new problems. Topology optimization (TO) - a well-established computational method for designing structures with optimized performance - lacks this ability to learn from experience. Existing approaches treat design tasks in isolation, starting from a "blank canvas" design for each new problem, often requiring many computationally expensive steps to converge. We propose a meta-learning strategy, termed meta-neural TO, that finds effective initial designs through a systematic transfer of knowledge between related tasks, building on the mesh-agnostic representation provided by neural reparameterization. We compare our approach against established TO methods, demonstrating efficient optimization across diverse test cases without compromising design quality. Further, we demonstrate powerful cross-resolution transfer capabilities, where initializations learned on lower-resolution discretizations lead to superior convergence in 74.1% of tasks on a higher-resolution test set, reducing the average number of iterations by 33.6% compared to standard neural TO. Remarkably, we discover that meta-learning naturally gravitates toward the strain energy patterns found in uniform density designs as effective starting points, aligning with engineering intuition.
2502.01834
Building a Cognitive Twin Using a Distributed Cognitive System and an Evolution Strategy
cs.AI cs.NE
This work presents a technique to build interaction-based Cognitive Twins (a computational version of an external agent) using input-output training and an Evolution Strategy on top of a framework for distributed Cognitive Architectures. Here, we show that it's possible to orchestrate many simple physical and virtual devices to achieve good approximations of a person's interaction behavior by training the system in an end-to-end fashion and present performance metrics. The generated Cognitive Twin may later be used to automate tasks, generate more realistic human-like artificial agents or further investigate its behaviors.
2502.01836
LeaFi: Data Series Indexes on Steroids with Learned Filters
cs.DB
The ever-growing collections of data series create a pressing need for efficient similarity search, which serves as the backbone for various analytics pipelines. Recent studies have shown that tree-based series indexes excel in many scenarios. However, we observe a significant waste of effort during search, due to suboptimal pruning. To address this issue, we introduce LeaFi, a novel framework that uses machine learning models to boost pruning effectiveness of tree-based data series indexes. These models act as learned filters, which predict tight node-wise distance lower bounds that are used to make pruning decisions, thus, improving pruning effectiveness. We describe the LeaFi-enhanced index building algorithm, which selects leaf nodes and generates training data to insert and train machine learning models, as well as the LeaFi-enhanced search algorithm, which calibrates learned filters at query time to support the user-defined quality target of each query. Our experimental evaluation, using two different tree-based series indexes and five diverse datasets, demonstrates the advantages of the proposed approach. LeaFi-enhanced data-series indexes improve pruning ratio by up to 20x and search time by up to 32x, while maintaining a target recall of 99%.
2502.01837
TESS: A Scalable Temporally and Spatially Local Learning Rule for Spiking Neural Networks
cs.NE cs.AI cs.LG
The demand for low-power inference and training of deep neural networks (DNNs) on edge devices has intensified the need for algorithms that are both scalable and energy-efficient. While spiking neural networks (SNNs) allow for efficient inference by processing complex spatio-temporal dynamics in an event-driven fashion, training them on resource-constrained devices remains challenging due to the high computational and memory demands of conventional error backpropagation (BP)-based approaches. In this work, we draw inspiration from biological mechanisms such as eligibility traces, spike-timing-dependent plasticity, and neural activity synchronization to introduce TESS, a temporally and spatially local learning rule for training SNNs. Our approach addresses both temporal and spatial credit assignments by relying solely on locally available signals within each neuron, thereby allowing computational and memory overheads to scale linearly with the number of neurons, independently of the number of time steps. Despite relying on local mechanisms, we demonstrate performance comparable to the backpropagation through time (BPTT) algorithm, within $\sim1.4$ accuracy points on challenging computer vision scenarios relevant at the edge, such as the IBM DVS Gesture dataset, CIFAR10-DVS, and temporal versions of CIFAR10, and CIFAR100. Being able to produce comparable performance to BPTT while keeping low time and memory complexity, TESS enables efficient and scalable on-device learning at the edge.
2502.01839
Sample, Scrutinize and Scale: Effective Inference-Time Search by Scaling Verification
cs.LG cs.AI
Sampling-based search, a simple paradigm for utilizing test-time compute, involves generating multiple candidate responses and selecting the best one -- typically by having models self-verify each response for correctness. In this paper, we study the scaling trends governing sampling-based search. Among our findings is that simply scaling up a minimalist implementation of sampling-based search, using only random sampling and direct self-verification, provides a practical inference method that, for example, elevates the reasoning capabilities of Gemini v1.5 Pro above that of o1-Preview on popular benchmarks. We partially attribute the scalability of sampling-based search to a phenomenon of implicit scaling, where sampling a larger pool of responses in turn improves self-verification accuracy. We further identify two useful principles for improving self-verification capabilities with test-time compute: (1) comparing across responses provides helpful signals about the locations of errors and hallucinations, and (2) different model output styles are useful for different contexts -- chains of thought are useful for reasoning but harder to verify. We also find that, though accurate verification can be elicited, frontier models demonstrate remarkably weak out-of-box verification capabilities and introduce a benchmark to measure progress on these deficiencies.
2502.01842
Texture Image Synthesis Using Spatial GAN Based on Vision Transformers
cs.CV cs.AI
Texture synthesis is a fundamental task in computer vision, whose goal is to generate visually realistic and structurally coherent textures for a wide range of applications, from graphics to scientific simulations. While traditional methods like tiling and patch-based techniques often struggle with complex textures, recent advancements in deep learning have transformed this field. In this paper, we propose ViT-SGAN, a new hybrid model that fuses Vision Transformers (ViTs) with a Spatial Generative Adversarial Network (SGAN) to address the limitations of previous methods. By incorporating specialized texture descriptors such as mean-variance (mu, sigma) and textons into the self-attention mechanism of ViTs, our model achieves superior texture synthesis. This approach enhances the model's capacity to capture complex spatial dependencies, leading to improved texture quality that is superior to state-of-the-art models, especially for regular and irregular textures. Comparison experiments with metrics such as FID, IS, SSIM, and LPIPS demonstrate the substantial improvement of ViT-SGAN, which underlines its efficiency in generating diverse realistic textures.
2502.01846
UVGS: Reimagining Unstructured 3D Gaussian Splatting using UV Mapping
cs.CV
3D Gaussian Splatting (3DGS) has demonstrated superior quality in modeling 3D objects and scenes. However, generating 3DGS remains challenging due to their discrete, unstructured, and permutation-invariant nature. In this work, we present a simple yet effective method to overcome these challenges. We utilize spherical mapping to transform 3DGS into a structured 2D representation, termed UVGS. UVGS can be viewed as multi-channel images, with feature dimensions as a concatenation of Gaussian attributes such as position, scale, color, opacity, and rotation. We further find that these heterogeneous features can be compressed into a lower-dimensional (e.g., 3-channel) shared feature space using a carefully designed multi-branch network. The compressed UVGS can be treated as typical RGB images. Remarkably, we discover that typical VAEs trained with latent diffusion models can directly generalize to this new representation without additional training. Our novel representation makes it effortless to leverage foundational 2D models, such as diffusion models, to directly model 3DGS. Additionally, one can simply increase the 2D UV resolution to accommodate more Gaussians, making UVGS a scalable solution compared to typical 3D backbones. This approach immediately unlocks various novel generation applications of 3DGS by inherently utilizing the already developed superior 2D generation capabilities. In our experiments, we demonstrate various unconditional, conditional generation, and inpainting applications of 3DGS based on diffusion models, which were previously non-trivial.
2502.01847
Containment Control Approach for Steering Opinion in a Social Network
eess.SY cs.MA cs.SY math.DS math.OC
The paper studies the problem of steering multi-dimensional opinion in a social network. Assuming the society of desire consists of stubborn and regular agents, stubborn agents are considered as leaders who specify the desired opinion distribution as a distributed reward or utility function. In this context, each regular agent is seen as a follower, updating its bias on the initial opinion and influence weights by averaging their observations of the rewards their influencers have received. Assuming random graphs with reducible and irreducible topology specify the influences on regular agents, opinion evolution is represented as a containment control problem in which stability and convergence to the final opinion are proven.
2502.01850
Foundation Model-Based Apple Ripeness and Size Estimation for Selective Harvesting
cs.CV
Harvesting is a critical task in the tree fruit industry, demanding extensive manual labor and substantial costs, and exposing workers to potential hazards. Recent advances in automated harvesting offer a promising solution by enabling efficient, cost-effective, and ergonomic fruit picking within tight harvesting windows. However, existing harvesting technologies often indiscriminately harvest all visible and accessible fruits, including those that are unripe or undersized. This study introduces a novel foundation model-based framework for efficient apple ripeness and size estimation. Specifically, we curated two public RGBD-based Fuji apple image datasets, integrating expanded annotations for ripeness ("Ripe" vs. "Unripe") based on fruit color and image capture dates. The resulting comprehensive dataset, Fuji-Ripeness-Size Dataset, includes 4,027 images and 16,257 annotated apples with ripeness and size labels. Using Grounding-DINO, a language-model-based object detector, we achieved robust apple detection and ripeness classification, outperforming other state-of-the-art models. Additionally, we developed and evaluated six size estimation algorithms, selecting the one with the lowest error and variation for optimal performance. The Fuji-Ripeness-Size Dataset and the apple detection and size estimation algorithms are made publicly available, which provides valuable benchmarks for future studies in automated and selective harvesting.
2502.01853
Security and Quality in LLM-Generated Code: A Multi-Language, Multi-Model Analysis
cs.CR cs.LG cs.SE
Artificial Intelligence (AI)-driven code generation tools are increasingly used throughout the software development lifecycle to accelerate coding tasks. However, the security of AI-generated code using Large Language Models (LLMs) remains underexplored, with studies revealing various risks and weaknesses. This paper analyzes the security of code generated by LLMs across different programming languages. We introduce a dataset of 200 tasks grouped into six categories to evaluate the performance of LLMs in generating secure and maintainable code. Our research shows that while LLMs can automate code creation, their security effectiveness varies by language. Many models fail to utilize modern security features in recent compiler and toolkit updates, such as Java 17. Moreover, outdated methods are still commonly used, particularly in C++. This highlights the need for advancing LLMs to enhance security and quality while incorporating emerging best practices in programming languages.
2502.01854
How to warm-start your unfolding network
cs.LG eess.IV eess.SP
We present a new ensemble framework for boosting the performance of overparameterized unfolding networks solving the compressed sensing problem. We combine a state-of-the-art overparameterized unfolding network with a continuation technique, to warm-start a crucial quantity of the said network's architecture; we coin the resulting continued network C-DEC. Moreover, for training and evaluating C-DEC, we incorporate the log-cosh loss function, which enjoys both linear and quadratic behavior. Finally, we numerically assess C-DEC's performance on real-world images. Results showcase that the combination of continuation with the overparameterized unfolded architecture, trained and evaluated with the chosen loss function, yields smoother loss landscapes and improved reconstruction and generalization performance of C-DEC, consistently for all datasets.
2502.01855
Learning Fine-to-Coarse Cuboid Shape Abstraction
cs.CV cs.GR
The abstraction of 3D objects with simple geometric primitives like cuboids allows to infer structural information from complex geometry. It is important for 3D shape understanding, structural analysis and geometric modeling. We introduce a novel fine-to-coarse unsupervised learning approach to abstract collections of 3D shapes. Our architectural design allows us to reduce the number of primitives from hundreds (fine reconstruction) to only a few (coarse abstraction) during training. This allows our network to optimize the reconstruction error and adhere to a user-specified number of primitives per shape while simultaneously learning a consistent structure across the whole collection of data. We achieve this through our abstraction loss formulation which increasingly penalizes redundant primitives. Furthermore, we introduce a reconstruction loss formulation to account not only for surface approximation but also volume preservation. Combining both contributions allows us to represent 3D shapes more precisely with fewer cuboid primitives than previous work. We evaluate our method on collections of man-made and humanoid shapes comparing with previous state-of-the-art learning methods on commonly used benchmarks. Our results confirm an improvement over previous cuboid-based shape abstraction techniques. Furthermore, we demonstrate our cuboid abstraction in downstream tasks like clustering, retrieval, and partial symmetry detection.
2502.01856
Reliability-Driven LiDAR-Camera Fusion for Robust 3D Object Detection
cs.CV cs.LG
Accurate and robust 3D object detection is essential for autonomous driving, where fusing data from sensors like LiDAR and camera enhances detection accuracy. However, sensor malfunctions such as corruption or disconnection can degrade performance, and existing fusion models often struggle to maintain reliability when one modality fails. To address this, we propose ReliFusion, a novel LiDAR-camera fusion framework operating in the bird's-eye view (BEV) space. ReliFusion integrates three key components: the Spatio-Temporal Feature Aggregation (STFA) module, which captures dependencies across frames to stabilize predictions over time; the Reliability module, which assigns confidence scores to quantify the dependability of each modality under challenging conditions; and the Confidence-Weighted Mutual Cross-Attention (CW-MCA) module, which dynamically balances information from LiDAR and camera modalities based on these confidence scores. Experiments on the nuScenes dataset show that ReliFusion significantly outperforms state-of-the-art methods, achieving superior robustness and accuracy in scenarios with limited LiDAR fields of view and severe sensor malfunctions.
2502.01857
Learning Human Perception Dynamics for Informative Robot Communication
cs.RO cs.AI
Human-robot cooperative navigation is challenging in environments with incomplete information. We introduce CoNav-Maze, a simulated robotics environment where a robot navigates using local perception while a human operator provides guidance based on an inaccurate map. The robot can share its camera views to improve the operator's understanding of the environment. To enable efficient human-robot cooperation, we propose Information Gain Monte Carlo Tree Search (IG-MCTS), an online planning algorithm that balances autonomous movement and informative communication. Central to IG-MCTS is a neural human perception dynamics model that estimates how humans distill information from robot communications. We collect a dataset through a crowdsourced mapping task in CoNav-Maze and train this model using a fully convolutional architecture with data augmentation. User studies show that IG-MCTS outperforms teleoperation and instruction-following baselines, achieving comparable task performance with significantly less communication and lower human cognitive load, as evidenced by eye-tracking metrics.
2502.01858
Rethinking Energy Management for Autonomous Ground Robots on a Budget
cs.RO cs.SY eess.SY
Autonomous Ground Robots (AGRs) face significant challenges due to limited energy reserve, which restricts their overall performance and availability. Prior research has focused separately on energy-efficient approaches and fleet management strategies for task allocation to extend operational time. A fleet-level scheduler, however, assumes a specific energy consumption during task allocation, requiring the AGR to fully utilize the energy for maximum performance, which contrasts with energy-efficient practices. This paper addresses this gap by investigating the combined impact of computing frequency and locomotion speed on energy consumption and performance. We analyze these variables through experiments on our prototype AGR, laying the foundation for an integrated approach that optimizes cyber-physical resources within the constraints of a specified energy budget. To tackle this challenge, we introduce PECC (Predictable Energy Consumption Controller), a framework designed to optimize computing frequency and locomotion speed to maximize performance while ensuring the system operates within the specified energy budget. We conducted extensive experiments with PECC using a real AGR and in simulations, comparing it to an energy-efficient baseline. Our results show that the AGR travels up to 17\% faster than the baseline in real-world tests and up to 31\% faster in simulations, while consuming 95\% and 91\% of the given energy budget, respectively. These results prove that PECC can effectively enhance AGR performance in scenarios where prioritizing the energy budget outweighs the need for energy efficiency.
2502.01860
SE Arena: Benchmarking Software Engineering Chatbots with Iterative Interactions
cs.SE cs.LG
Foundation models (FMs), particularly large language models (LLMs), have shown significant promise in various software engineering (SE) tasks, including code generation, debugging, and requirement refinement. Despite these advances, existing evaluation frameworks are insufficient for assessing model performance in iterative, context-rich workflows characteristic of SE activities. To address this limitation, we introduce SE Arena, an interactive platform designed to evaluate SE-focused chatbots. SE Arena provides a transparent, open-source leaderboard, supports multi-round conversational workflows, and enables end-to-end model comparisons. Moreover, SE Arena incorporates a new feature called RepoChat, which automatically injects repository-related context (e.g., issues, commits, pull requests) into the conversation, further aligning evaluations with real-world development processes. This paper outlines the design and capabilities of SE Arena, emphasizing its potential to advance the evaluation and practical application of FMs in software engineering.
2502.01861
Learning Hyperparameters via a Data-Emphasized Variational Objective
cs.LG stat.ML
When training large flexible models, practitioners often rely on grid search to select hyperparameters that control over-fitting. This grid search has several disadvantages: the search is computationally expensive, requires carving out a validation set that reduces the available data for training, and requires users to specify candidate values. In this paper, we propose an alternative: directly learning regularization hyperparameters on the full training set via the evidence lower bound ("ELBo") objective from variational methods. For deep neural networks with millions of parameters, we recommend a modified ELBo that upweights the influence of the data likelihood relative to the prior. Our proposed technique overcomes all three disadvantages of grid search. In a case study on transfer learning of image classifiers, we show how our method reduces the 88+ hour grid search of past work to under 3 hours while delivering comparable accuracy. We further demonstrate how our approach enables efficient yet accurate approximations of Gaussian processes with learnable length-scale kernels.
2502.01865
Enhancing Generalization via Sharpness-Aware Trajectory Matching for Dataset Condensation
cs.LG
Dataset condensation aims to synthesize datasets with a few representative samples that can effectively represent the original datasets. This enables efficient training and produces models with performance close to those trained on the original sets. Most existing dataset condensation methods conduct dataset learning under the bilevel (inner- and outer-loop) based optimization. However, the preceding methods perform with limited dataset generalization due to the notoriously complicated loss landscape and expensive time-space complexity of the inner-loop unrolling of bilevel optimization. These issues deteriorate when the datasets are learned via matching the trajectories of networks trained on the real and synthetic datasets with a long horizon inner-loop. To address these issues, we introduce Sharpness-Aware Trajectory Matching (SATM), which enhances the generalization capability of learned synthetic datasets by optimising the sharpness of the loss landscape and objective simultaneously. Moreover, our approach is coupled with an efficient hypergradient approximation that is mathematically well-supported and straightforward to implement along with controllable computational overhead. Empirical evaluations of SATM demonstrate its effectiveness across various applications, including in-domain benchmarks and out-of-domain settings. Moreover, its easy-to-implement properties afford flexibility, allowing it to integrate with other advanced sharpness-aware minimizers. Our code will be released.
2502.01866
Online Curvature-Aware Replay: Leveraging $\mathbf{2^{nd}}$ Order Information for Online Continual Learning
cs.LG cs.AI
Online Continual Learning (OCL) models continuously adapt to nonstationary data streams, usually without task information. These settings are complex and many traditional CL methods fail, while online methods (mainly replay-based) suffer from instabilities after the task shift. To address this issue, we formalize replay-based OCL as a second-order online joint optimization with explicit KL-divergence constraints on replay data. We propose Online Curvature-Aware Replay (OCAR) to solve the problem: a method that leverages second-order information of the loss using a K-FAC approximation of the Fisher Information Matrix (FIM) to precondition the gradient. The FIM acts as a stabilizer to prevent forgetting while also accelerating the optimization in non-interfering directions. We show how to adapt the estimation of the FIM to a continual setting stabilizing second-order optimization for non-iid data, uncovering the role of the Tikhonov regularization in the stability-plasticity tradeoff. Empirical results show that OCAR outperforms state-of-the-art methods in continual metrics achieving higher average accuracy throughout the training process in three different benchmarks.
2502.01867
Optimizing Online Advertising with Multi-Armed Bandits: Mitigating the Cold Start Problem under Auction Dynamics
cs.LG
Online advertising platforms often face a common challenge: the cold start problem. Insufficient behavioral data (clicks) makes accurate click-through rate (CTR) forecasting of new ads challenging. CTR for "old" items can also be significantly underestimated due to their early performance influencing their long-term behavior on the platform. The cold start problem has far-reaching implications for businesses, including missed long-term revenue opportunities. To mitigate this issue, we developed a UCB-like algorithm under multi-armed bandit (MAB) setting for positional-based model (PBM), specifically tailored to auction pay-per-click systems. Our proposed algorithm successfully combines theory and practice: we obtain theoretical upper estimates of budget regret, and conduct a series of experiments on synthetic and real-world data that confirm the applicability of the method on the real platform. In addition to increasing the platform's long-term profitability, we also propose a mechanism for maintaining short-term profits through controlled exploration and exploitation of items.
2502.01873
Explaining Automatic Image Assessment
cs.CV
Previous work in aesthetic categorization and explainability utilizes manual labeling and classification to explain aesthetic scores. These methods require a complex labeling process and are limited in size. Our proposed approach attempts to explain aesthetic assessment models through visualizing dataset trends and automatic categorization of visual aesthetic features through training neural networks on different versions of the same dataset. By evaluating the models adapted to each specific modality using existing and novel metrics, we can capture and visualize aesthetic features and trends.
2502.01874
Countering Election Sway: Strategic Algorithms in Friedkin-Johnsen Dynamics
cs.SI
Social influence profoundly impacts individual choices and collective behaviors in politics. In this work, driven by the goal of protecting elections from improper influence, we consider the following scenario: an individual, who has vested interests in political party $Y$, is aware through reliable surveys that parties $X$ and $Y$ are likely to get 50.1\% and 49.9\% of the vote, respectively. Could this individual employ strategies to alter public opinions and consequently invert these polling numbers in favor of party $Y$? We address this question by employing: (i) the Friedkin-Johnsen (FJ) opinion dynamics model, which is mathematically sophisticated and effectively captures the way individual biases and social interactions shape opinions, making it crucial for examining social influence, and (ii) interventions similar to those in Asch's experiments, which involve selecting a group of stooges within the network to spread a specific opinion. We mathematically formalize the aforementioned motivation as an optimization framework and establish that it is NP-hard and inapproximable within any constant factor. We introduce three efficient polynomial-time algorithms. The first two utilize a continuous approach: one employs gradient descent with Huber's estimator to approximate the median, and the other uses a sigmoid threshold influence function. The third utilizes a combinatorial greedy algorithm for targeted interventions. Through comparative analysis against various natural baselines and using real-world data, our results demonstrate that in numerous cases a small fraction of nodes chosen as stooges can significantly sway election outcomes under the Friedkin-Johnsen model.
2502.01876
Reinforcement Learning with Segment Feedback
cs.LG
Standard reinforcement learning (RL) assumes that an agent can observe a reward for each state-action pair. However, in practical applications, it is often difficult and costly to collect a reward for each state-action pair. While there have been several works considering RL with trajectory feedback, it is unclear if trajectory feedback is inefficient for learning when trajectories are long. In this work, we consider a model named RL with segment feedback, which offers a general paradigm filling the gap between per-state-action feedback and trajectory feedback. In this model, we consider an episodic Markov decision process (MDP), where each episode is divided into $m$ segments, and the agent observes reward feedback only at the end of each segment. Under this model, we study two popular feedback settings: binary feedback and sum feedback, where the agent observes a binary outcome and a reward sum according to the underlying reward function, respectively. To investigate the impact of the number of segments $m$ on learning performance, we design efficient algorithms and establish regret upper and lower bounds for both feedback settings. Our theoretical and experimental results show that: under binary feedback, increasing the number of segments $m$ decreases the regret at an exponential rate; in contrast, surprisingly, under sum feedback, increasing $m$ does not reduce the regret significantly.
2502.01882
Latent Lexical Projection in Large Language Models: A Novel Approach to Implicit Representation Refinement
cs.CL
Generating semantically coherent text requires a robust internal representation of linguistic structures, which traditional embedding techniques often fail to capture adequately. A novel approach, Latent Lexical Projection (LLP), is introduced to refine lexical representations through a structured transformation into a latent space, thereby enhancing the alignment between input embeddings and their contextual meanings. The method integrates an optimized projection mechanism within an existing language model architecture, enabling more accurate token selection while maintaining syntactic integrity. Evaluations across multiple benchmarks indicate a reduction in perplexity and an increase in BLEU scores, suggesting improvements in predictive accuracy and fluency. The analysis of lexical diversity reveals a more varied vocabulary in generated text, addressing common issues of redundancy and repetitive phrase structures. Further assessments of entropy distributions demonstrate a decline in uncertainty during decoding, reflecting enhanced confidence in word selection. Additionally, long-range dependency retention exhibits measurable gains, with increased classification accuracy at extended token distances. Computational efficiency remains within manageable constraints, despite the added projection mechanism, highlighting the practicality of LLP for integration into existing architectures.
2502.01885
A Privacy-Preserving Domain Adversarial Federated learning for multi-site brain functional connectivity analysis
cs.LG cs.AI eess.IV
Resting-state functional magnetic resonance imaging (rs-fMRI) and its derived functional connectivity networks (FCNs) have become critical for understanding neurological disorders. However, collaborative analyses and the generalizability of models still face significant challenges due to privacy regulations and the non-IID (non-independent and identically distributed) property of multiple data sources. To mitigate these difficulties, we propose Domain Adversarial Federated Learning (DAFed), a novel federated deep learning framework specifically designed for non-IID fMRI data analysis in multi-site settings. DAFed addresses these challenges through feature disentanglement, decomposing the latent feature space into domain-invariant and domain-specific components, to ensure robust global learning while preserving local data specificity. Furthermore, adversarial training facilitates effective knowledge transfer between labeled and unlabeled datasets, while a contrastive learning module enhances the global representation of domain-invariant features. We evaluated DAFed on the diagnosis of ASD and further validated its generalizability in the classification of AD, demonstrating its superior classification accuracy compared to state-of-the-art methods. Additionally, an enhanced Score-CAM module identifies key brain regions and functional connectivity significantly associated with ASD and MCI, respectively, uncovering shared neurobiological patterns across sites. These findings highlight the potential of DAFed to advance multi-site collaborative research in neuroimaging while protecting data confidentiality.
2502.01889
Displacement-Sparse Neural Optimal Transport
cs.LG cs.AI
Optimal Transport (OT) theory seeks to determine the map $T:X \to Y$ that transports a source measure $P$ to a target measure $Q$, minimizing the cost $c(\mathbf{x}, T(\mathbf{x}))$ between $\mathbf{x}$ and its image $T(\mathbf{x})$. Building upon the Input Convex Neural Network OT solver and incorporating the concept of displacement-sparse maps, we introduce a sparsity penalty into the minimax Wasserstein formulation, promote sparsity in displacement vectors $\Delta(\mathbf{x}) := T(\mathbf{x}) - \mathbf{x}$, and enhance the interpretability of the resulting map. However, increasing sparsity often reduces feasibility, causing $T_{\#}(P)$ to deviate more significantly from the target measure. In low-dimensional settings, we propose a heuristic framework to balance the trade-off between sparsity and feasibility by dynamically adjusting the sparsity intensity parameter during training. For high-dimensional settings, we directly constrain the dimensionality of displacement vectors by enforcing $\dim(\Delta(\mathbf{x})) \leq l$, where $l < d$ for $X \subseteq \mathbb{R}^d$. Among maps satisfying this constraint, we aim to identify the most feasible one. This goal can be effectively achieved by adapting our low-dimensional heuristic framework without resorting to dimensionality reduction. We validate our method on both synthesized sc-RNA and real 4i cell perturbation datasets, demonstrating improvements over existing methods.
2502.01890
Geometric Framework for 3D Cell Segmentation Correction
cs.CV cs.LG
3D cellular image segmentation methods are commonly divided into non-2D-based and 2D-based approaches, the latter reconstructing 3D shapes from the segmentation results of 2D layers. However, errors in 2D results often propagate, leading to oversegmentations in the final 3D results. To tackle this issue, we introduce an interpretable geometric framework that addresses the oversegmentations by correcting the 2D segmentation results based on geometric information from adjacent layers. Leveraging both geometric (layer-to-layer, 2D) and topological (3D shape) features, we use binary classification to determine whether neighboring cells should be stitched. We develop a pre-trained classifier on public plant cell datasets and validate its performance on animal cell datasets, confirming its effectiveness in correcting oversegmentations under the transfer learning setting. Furthermore, we demonstrate that our framework can be extended to correcting oversegmentation on non-2D-based methods. A clear pipeline is provided for end-users to build the pre-trained model to any labeled dataset.
2502.01891
Training and Evaluating with Human Label Variation: An Empirical Study
cs.LG cs.CL
Human label variation (HLV) challenges the standard assumption that an example has a single ground truth, instead embracing the natural variation in human labelling to train and evaluate models. While various training methods and metrics for HLV have been proposed, there has been no systematic meta-evaluation of HLV evaluation metrics, contributing to the lack of clarity in the best HLV training method. We propose new evaluation metrics and training methods and empirically meta-evaluate HLV evaluation metrics. We find that training on either disaggregated annotations or soft labels often performs best across metrics, and that our proposed soft metric correlates best with human preference.
2502.01894
SimBEV: A Synthetic Multi-Task Multi-Sensor Driving Data Generation Tool and Dataset
cs.CV cs.LG cs.RO
Bird's-eye view (BEV) perception for autonomous driving has garnered significant attention in recent years, in part because BEV representation facilitates the fusion of multi-sensor data. This enables a variety of perception tasks including BEV segmentation, a concise view of the environment that can be used to plan a vehicle's trajectory. However, this representation is not fully supported by existing datasets, and creation of new datasets can be a time-consuming endeavor. To address this problem, in this paper we introduce SimBEV, an extensively configurable and scalable randomized synthetic data generation tool that incorporates information from multiple sources to capture accurate BEV ground truth data, supports a comprehensive array of sensors, and enables a variety of perception tasks including BEV segmentation and 3D object detection. We use SimBEV to create the SimBEV dataset, a large collection of annotated perception data from diverse driving scenarios.
2502.01896
INTACT: Inducing Noise Tolerance through Adversarial Curriculum Training for LiDAR-based Safety-Critical Perception and Autonomy
cs.CV cs.RO
In this work, we present INTACT, a novel two-phase framework designed to enhance the robustness of deep neural networks (DNNs) against noisy LiDAR data in safety-critical perception tasks. INTACT combines meta-learning with adversarial curriculum training (ACT) to systematically address challenges posed by data corruption and sparsity in 3D point clouds. The meta-learning phase equips a teacher network with task-agnostic priors, enabling it to generate robust saliency maps that identify critical data regions. The ACT phase leverages these saliency maps to progressively expose a student network to increasingly complex noise patterns, ensuring targeted perturbation and improved noise resilience. INTACT's effectiveness is demonstrated through comprehensive evaluations on object detection, tracking, and classification benchmarks using diverse datasets, including KITTI, Argoverse, and ModelNet40. Results indicate that INTACT improves model robustness by up to 20% across all tasks, outperforming standard adversarial and curriculum training methods. This framework not only addresses the limitations of conventional training strategies but also offers a scalable and efficient solution for real-world deployment in resource-constrained safety-critical systems. INTACT's principled integration of meta-learning and adversarial training establishes a new paradigm for noise-tolerant 3D perception in safety-critical applications. INTACT improved KITTI Multiple Object Tracking Accuracy (MOTA) by 9.6% (64.1% -> 75.1%) and by 12.4% under Gaussian noise (52.5% -> 73.7%). Similarly, KITTI mean Average Precision (mAP) rose from 59.8% to 69.8% (50% point drop) and 49.3% to 70.9% (Gaussian noise), highlighting the framework's ability to enhance deep learning model resilience in safety-critical object tracking scenarios.
2502.01901
Conceptual Metaphor Theory as a Prompting Paradigm for Large Language Models
cs.CL
We introduce Conceptual Metaphor Theory (CMT) as a framework for enhancing large language models (LLMs) through cognitive prompting in complex reasoning tasks. CMT leverages metaphorical mappings to structure abstract reasoning, improving models' ability to process and explain intricate concepts. By incorporating CMT-based prompts, we guide LLMs toward more structured and human-like reasoning patterns. To evaluate this approach, we compare four native models (Llama3.2, Phi3, Gemma2, and Mistral) against their CMT-augmented counterparts on benchmark tasks spanning domain-specific reasoning, creative insight, and metaphor interpretation. Responses were automatically evaluated using the Llama3.3 70B model. Experimental results indicate that CMT prompting significantly enhances reasoning accuracy, clarity, and metaphorical coherence, outperforming baseline models across all evaluated tasks.
2502.01904
Common Neighborhood Estimation over Bipartite Graphs under Local Differential Privacy
cs.DB
Bipartite graphs, formed by two vertex layers, arise as a natural fit for modeling the relationships between two groups of entities. In bipartite graphs, common neighborhood computation between two vertices on the same vertex layer is a basic operator, which is easily solvable in general settings. However, it inevitably involves releasing the neighborhood information of vertices, posing a significant privacy risk for users in real-world applications. To protect edge privacy in bipartite graphs, in this paper, we study the problem of estimating the number of common neighbors of two vertices on the same layer under edge local differential privacy (edge LDP). The problem is challenging in the context of edge LDP since each vertex on the opposite layer of the query vertices can potentially be a common neighbor. To obtain efficient and accurate estimates, we propose a multiple-round framework that significantly reduces the candidate pool of common neighbors and enables the query vertices to construct unbiased estimators locally. Furthermore, we improve data utility by incorporating the estimators built from the neighbors of both query vertices and devise privacy budget allocation optimizations. These improve the estimator's robustness and consistency, particularly against query vertices with imbalanced degrees. Extensive experiments on 15 datasets validate the effectiveness and efficiency of our proposed techniques.
2502.01905
When not to target negative ties? Studying competitive influence maximisation in signed networks
cs.SI
We explore the influence maximisation problem in networks with negative ties. Where prior work has focused on unsigned networks, we investigate the need to consider negative ties in networks while trying to maximise spread in a population - particularly under competitive conditions. Given a signed network we optimise the strategies of a focal controller, against competing influence in the network, using two approaches - either the focal controller uses a sign-agnostic approach or they factor in the sign of the edges while optimising their strategy. We compare the difference in vote-shares (or the share of population) obtained by both these methods to determine the need to navigate negative ties in these settings. More specifically, we study the impact of: (a) network topology, (b) resource conditions and (c) competitor strategies on the difference in vote shares obtained across both methodologies. We observe that gains are maximum when resources available to the focal controller are low and the competitor avoids negative edges in their strategy. Conversely, gains are insignificant irrespective of resource conditions when the competitor targets the network indiscriminately. Finally, we study the problem in a game-theoretic setting, where we simultaneously optimise the strategies of both competitors. Interestingly we observe that, strategising with the knowledge of negative ties can occasionally also lead to loss in vote-shares.
2502.01906
Rethinking Homogeneity of Vision and Text Tokens in Large Vision-and-Language Models
cs.CV
Large vision-and-language models (LVLMs) typically treat visual and textual embeddings as homogeneous inputs to a large language model (LLM). However, these inputs are inherently different: visual inputs are multi-dimensional and contextually rich, often pre-encoded by models like CLIP, while textual inputs lack this structure. In this paper, we propose Decomposed Attention (D-Attn), a novel method that processes visual and textual embeddings differently by decomposing the 1-D causal self-attention in LVLMs. After the attention decomposition, D-Attn diagonalizes visual-to-visual self-attention, reducing computation from $\mathcal{O}(|V|^2)$ to $\mathcal{O}(|V|)$ for $|V|$ visual embeddings without compromising performance. Moreover, D-Attn debiases positional encodings in textual-to-visual cross-attention, further enhancing visual understanding. Finally, we introduce an $\alpha$-weighting strategy to merge visual and textual information, maximally preserving the pre-trained LLM's capabilities with minimal modifications. Extensive experiments and rigorous analyses validate the effectiveness of D-Attn, demonstrating significant improvements on multiple image benchmarks while significantly reducing computational costs. Code, data, and models will be publicly available.
2502.01908
Unlocking Efficient Large Inference Models: One-Bit Unrolling Tips the Scales
cs.LG
Recent advancements in Large Language Model (LLM) compression, such as BitNet and BitNet b1.58, have marked significant strides in reducing the computational demands of LLMs through innovative one-bit quantization techniques. We extend this frontier by looking at Large Inference Models (LIMs) that have become indispensable across various applications. However, their scale and complexity often come at a significant computational cost. We introduce a novel approach that leverages one-bit algorithm unrolling, effectively integrating information from the physical world in the model architecture. Our method achieves a bit-per-link rate significantly lower than the 1.58 bits reported in prior work, thanks to the natural sparsity that emerges in our network architectures. We numerically demonstrate that the proposed one-bit algorithm unrolling scheme can improve both training and test outcomes by effortlessly increasing the number of layers while substantially compressing the network. Additionally, we provide theoretical results on the generalization gap, convergence rate, stability, and sensitivity of our proposed one-bit algorithm unrolling.
2502.01912
PATCH: a deep learning method to assess heterogeneity of artistic practice in historical paintings
cs.CV cs.AI cs.LG
The history of art has seen significant shifts in the manner in which artworks are created, making understanding of creative processes a central question in technical art history. In the Renaissance and Early Modern period, paintings were largely produced by master painters directing workshops of apprentices who often contributed to projects. The masters varied significantly in artistic and managerial styles, meaning different combinations of artists and implements might be seen both between masters and within workshops or even individual canvases. Information on how different workshops were managed and the processes by which artworks were created remains elusive. Machine learning methods have potential to unearth new information about artists' creative processes by extending the analysis of brushwork to a microscopic scale. Analysis of workshop paintings, however, presents a challenge in that documentation of the artists and materials involved is sparse, meaning external examples are not available to train networks to recognize their contributions. Here we present a novel machine learning approach we call pairwise assignment training for classifying heterogeneity (PATCH) that is capable of identifying individual artistic practice regimes with no external training data, or "ground truth." The method achieves unsupervised results by supervised means, and outperforms both simple statistical procedures and unsupervised machine learning methods. We apply this method to two historical paintings by the Spanish Renaissance master, El Greco: The Baptism of Christ and Christ on the Cross with Landscape, and our findings regarding the former potentially challenge previous work that has assigned the painting to workshop members. Further, the results of our analyses create a measure of heterogeneity of artistic practice that can be used to characterize artworks across time and space.
2502.01913
Composite Gaussian Processes Flows for Learning Discontinuous Multimodal Policies
cs.RO cs.LG
Learning control policies for real-world robotic tasks often involve challenges such as multimodality, local discontinuities, and the need for computational efficiency. These challenges arise from the complexity of robotic environments, where multiple solutions may coexist. To address these issues, we propose Composite Gaussian Processes Flows (CGP-Flows), a novel semi-parametric model for robotic policy. CGP-Flows integrate Overlapping Mixtures of Gaussian Processes (OMGPs) with the Continuous Normalizing Flows (CNFs), enabling them to model complex policies addressing multimodality and local discontinuities. This hybrid approach retains the computational efficiency of OMGPs while incorporating the flexibility of CNFs. Experiments conducted in both simulated and real-world robotic tasks demonstrate that CGP-flows significantly improve performance in modeling control policies. In a simulation task, we confirmed that CGP-Flows had a higher success rate compared to the baseline method, and the success rate of GCP-Flow was significantly different from the success rate of other baselines in chi-square tests.
2502.01916
Generalizable and Fast Surrogates: Model Predictive Control of Articulated Soft Robots using Physics-Informed Neural Networks
cs.RO cs.LG
Soft robots can revolutionize several applications with high demands on dexterity and safety. When operating these systems, real-time estimation and control require fast and accurate models. However, prediction with first-principles (FP) models is slow, and learned black-box models have poor generalizability. Physics-informed machine learning offers excellent advantages here, but it is currently limited to simple, often simulated systems without considering changes after training. We propose physics-informed neural networks (PINNs) for articulated soft robots (ASRs) with a focus on data efficiency. The amount of expensive real-world training data is reduced to a minimum - one dataset in one system domain. Two hours of data in different domains are used for a comparison against two gold-standard approaches: In contrast to a recurrent neural network, the PINN provides a high generalizability. The prediction speed of an accurate FP model is improved with the PINN by up to a factor of 466 at slightly reduced accuracy. This enables nonlinear model predictive control (MPC) of the pneumatic ASR. In nine dynamic MPC experiments, an average joint-tracking error of 1.3{\deg} is achieved.
2502.01918
Wake-Informed 3D Path Planning for Autonomous Underwater Vehicles Using A* and Neural Network Approximations
cs.RO cs.AI cs.LG
Autonomous Underwater Vehicles (AUVs) encounter significant energy, control and navigation challenges in complex underwater environments, particularly during close-proximity operations, such as launch and recovery (LAR), where fluid interactions and wake effects present additional navigational and energy challenges. Traditional path planning methods fail to incorporate these detailed wake structures, resulting in increased energy consumption, reduced control stability, and heightened safety risks. This paper presents a novel wake-informed, 3D path planning approach that fully integrates localized wake effects and global currents into the planning algorithm. Two variants of the A* algorithm - a current-informed planner and a wake-informed planner - are created to assess its validity and two neural network models are then trained to approximate these planners for real-time applications. Both the A* planners and NN models are evaluated using important metrics such as energy expenditure, path length, and encounters with high-velocity and turbulent regions. The results demonstrate a wake-informed A* planner consistently achieves the lowest energy expenditure and minimizes encounters with high-velocity regions, reducing energy consumption by up to 11.3%. The neural network models are observed to offer computational speedup of 6 orders of magnitude, but exhibit 4.51 - 19.79% higher energy expenditures and 9.81 - 24.38% less optimal paths. These findings underscore the importance of incorporating detailed wake structures into traditional path planning algorithms and the benefits of neural network approximations to enhance energy efficiency and operational safety for AUVs in complex 3D domains.
2502.01919
Poisson Hierarchical Indian Buffet Processes for Within and Across Group Sharing of Latent Features-With Indications for Microbiome Species Sampling Models
stat.ML cs.LG math.PR math.ST stat.TH
In this work, we present a comprehensive Bayesian posterior analysis of what we term Poisson Hierarchical Indian Buffet Processes, designed for complex random sparse count species sampling models that allow for the sharing of information across and within groups. This analysis covers a potentially infinite number of species and unknown parameters, which, within a Bayesian machine learning context, we are able to learn from as more information is sampled. To achieve our refined results, we employ a range of methodologies drawn from Bayesian latent feature models, random occupancy models, and excursion theory. Despite this complexity, our goal is to make our findings accessible to practitioners, including those who may not be familiar with these areas. To facilitate understanding, we adopt a pseudo-expository style that emphasizes clarity and practical utility. We aim to express our findings in a language that resonates with experts in microbiome and ecological studies, addressing gaps in modeling capabilities while acknowledging that we are not experts ourselves in these fields. This approach encourages the use of our models as basic components of more sophisticated frameworks employed by domain experts, embodying the spirit of the seminal work on the Dirichlet Process. Ultimately, our refined posterior analysis not only yields tractable computational procedures but also enables practical statistical implementation and provides a clear mapping to relevant quantities in microbiome analysis.
2502.01920
Anomaly Detection via Autoencoder Composite Features and NCE
cs.LG
Unsupervised anomaly detection is a challenging task. Autoencoders (AEs) or generative models are often employed to model the data distribution of normal inputs and subsequently identify anomalous, out-of-distribution inputs by high reconstruction error or low likelihood, respectively. However, AEs may generalize and achieve small reconstruction errors on abnormal inputs. We propose a decoupled training approach for anomaly detection that both an AE and a likelihood model trained with noise contrastive estimation (NCE). After training the AE, NCE estimates a probability density function, to serve as the anomaly score, on the joint space of the AE's latent representation combined with features of the reconstruction quality. To further reduce the false negative rate in NCE we systematically varying the reconstruction features to augment the training and optimize the contrastive Gaussian noise distribution. Experimental assessments on multiple benchmark datasets demonstrate that the proposed approach matches the performance of prevalent state-of-the-art anomaly detection algorithms.
2502.01922
LAST SToP For Modeling Asynchronous Time Series
cs.LG cs.AI
We present a novel prompt design for Large Language Models (LLMs) tailored to Asynchronous Time Series. Unlike regular time series, which assume values at evenly spaced time points, asynchronous time series consist of timestamped events occurring at irregular intervals, each described in natural language. Our approach effectively utilizes the rich natural language of event descriptions, allowing LLMs to benefit from their broad world knowledge for reasoning across different domains and tasks. This allows us to extend the scope of asynchronous time series analysis beyond forecasting to include tasks like anomaly detection and data imputation. We further introduce Stochastic Soft Prompting, a novel prompt-tuning mechanism that significantly improves model performance, outperforming existing fine-tuning methods such as QLoRA. Through extensive experiments on real world datasets, we demonstrate that our approach achieves state-of-the-art performance across different tasks and datasets.
2502.01924
DualGuard MPPI: Safe and Performant Optimal Control by Combining Sampling-Based MPC and Hamilton-Jacobi Reachability
eess.SY cs.RO cs.SY
Designing controllers that are both safe and performant is inherently challenging. This co-optimization can be formulated as a constrained optimal control problem, where the cost function represents the performance criterion and safety is specified as a constraint. While sampling-based methods, such as Model Predictive Path Integral (MPPI) control, have shown great promise in tackling complex optimal control problems, they often struggle to enforce safety constraints. To address this limitation, we propose DualGuard-MPPI, a novel framework for solving safety-constrained optimal control problems. Our approach integrates Hamilton-Jacobi reachability analysis within the MPPI sampling process to ensure that all generated samples are provably safe for the system. On the one hand, this integration allows DualGuard-MPPI to enforce strict safety constraints; at the same time, it facilitates a more effective exploration of the environment with the same number of samples, reducing the effective sampling variance and leading to better performance optimization. Through several simulations and hardware experiments, we demonstrate that the proposed approach achieves much higher performance compared to existing MPPI methods, without compromising safety.
2502.01925
PANDAS: Improving Many-shot Jailbreaking via Positive Affirmation, Negative Demonstration, and Adaptive Sampling
cs.CL cs.CR cs.LG
Many-shot jailbreaking circumvents the safety alignment of large language models by exploiting their ability to process long input sequences. To achieve this, the malicious target prompt is prefixed with hundreds of fabricated conversational turns between the user and the model. These fabricated exchanges are randomly sampled from a pool of malicious questions and responses, making it appear as though the model has already complied with harmful instructions. In this paper, we present PANDAS: a hybrid technique that improves many-shot jailbreaking by modifying these fabricated dialogues with positive affirmations, negative demonstrations, and an optimized adaptive sampling method tailored to the target prompt's topic. Extensive experiments on AdvBench and HarmBench, using state-of-the-art LLMs, demonstrate that PANDAS significantly outperforms baseline methods in long-context scenarios. Through an attention analysis, we provide insights on how long-context vulnerabilities are exploited and show how PANDAS further improves upon many-shot jailbreaking.
2502.01926
Fairness through Difference Awareness: Measuring Desired Group Discrimination in LLMs
cs.CY cs.CL
Algorithmic fairness has conventionally adopted a perspective of racial color-blindness (i.e., difference unaware treatment). We contend that in a range of important settings, group difference awareness matters. For example, differentiating between groups may be necessary in legal contexts (e.g., the U.S. compulsory draft applies to men but not women) and harm assessments (e.g., calling a girl a terrorist may be less harmful than calling a Muslim person one). In our work we first introduce an important distinction between descriptive (fact-based), normative (value-based), and correlation (association-based) benchmarks. This distinction is significant because each category requires distinct interpretation and mitigation tailored to its specific characteristics. Then, we present a benchmark suite composed of eight different scenarios for a total of 16k questions that enables us to assess difference awareness. Finally, we show results across ten models that demonstrate difference awareness is a distinct dimension of fairness where existing bias mitigation strategies may backfire.
2502.01930
Distributionally Robust Direct Preference Optimization
cs.LG cs.AI
A major challenge in aligning large language models (LLMs) with human preferences is the issue of distribution shift. LLM alignment algorithms rely on static preference datasets, assuming that they accurately represent real-world user preferences. However, user preferences vary significantly across geographical regions, demographics, linguistic patterns, and evolving cultural trends. This preference distribution shift leads to catastrophic alignment failures in many real-world applications. We address this problem using the principled framework of distributionally robust optimization, and develop two novel distributionally robust direct preference optimization (DPO) algorithms, namely, Wasserstein DPO (WDPO) and Kullback-Leibler DPO (KLDPO). We characterize the sample complexity of learning the optimal policy parameters for WDPO and KLDPO. Moreover, we propose scalable gradient descent-style learning algorithms by developing suitable approximations for the challenging minimax loss functions of WDPO and KLDPO. Our empirical experiments demonstrate the superior performance of WDPO and KLDPO in substantially improving the alignment when there is a preference distribution shift.
2502.01932
VolleyBots: A Testbed for Multi-Drone Volleyball Game Combining Motion Control and Strategic Play
cs.RO cs.AI cs.LG
Multi-agent reinforcement learning (MARL) has made significant progress, largely fueled by the development of specialized testbeds that enable systematic evaluation of algorithms in controlled yet challenging scenarios. However, existing testbeds often focus on purely virtual simulations or limited robot morphologies such as robotic arms, quadrupeds, and humanoids, leaving high-mobility platforms with real-world physical constraints like drones underexplored. To bridge this gap, we present VolleyBots, a new MARL testbed where multiple drones cooperate and compete in the sport of volleyball under physical dynamics. VolleyBots features a turn-based interaction model under volleyball rules, a hierarchical decision-making process that combines motion control and strategic play, and a high-fidelity simulation for seamless sim-to-real transfer. We provide a comprehensive suite of tasks ranging from single-drone drills to multi-drone cooperative and competitive tasks, accompanied by baseline evaluations of representative MARL and game-theoretic algorithms. Results in simulation show that while existing algorithms handle simple tasks effectively, they encounter difficulty in complex tasks that require both low-level control and high-level strategy. We further demonstrate zero-shot deployment of a simulation-learned policy to real-world drones, highlighting VolleyBots' potential to propel MARL research involving agile robotic platforms. The project page is at https://sites.google.com/view/thu-volleybots/home.
2502.01936
Query-Based and Unnoticeable Graph Injection Attack from Neighborhood Perspective
cs.LG cs.CR
The robustness of Graph Neural Networks (GNNs) has become an increasingly important topic due to their expanding range of applications. Various attack methods have been proposed to explore the vulnerabilities of GNNs, ranging from Graph Modification Attacks (GMA) to the more practical and flexible Graph Injection Attacks (GIA). However, existing methods face two key challenges: (i) their reliance on surrogate models, which often leads to reduced attack effectiveness due to structural differences and prior biases, and (ii) existing GIA methods often sacrifice attack success rates in undefended settings to bypass certain defense models, thereby limiting their overall effectiveness. To overcome these limitations, we propose QUGIA, a Query-based and Unnoticeable Graph Injection Attack. QUGIA injects nodes by first selecting edges based on victim node connections and then generating node features using a Bayesian framework. This ensures that the injected nodes are similar to the original graph nodes, implicitly preserving homophily and making the attack more unnoticeable. Unlike previous methods, QUGIA does not rely on surrogate models, thereby avoiding performance degradation and achieving better generalization. Extensive experiments on six real-world datasets with diverse characteristics demonstrate that QUGIA achieves unnoticeable attacks and outperforms state-of-the-art attackers. The code will be released upon acceptance.
2502.01937
A Comprehensive Study of Bug-Fix Patterns in Autonomous Driving Systems
cs.SE cs.RO
As autonomous driving systems (ADSes) become increasingly complex and integral to daily life, the importance of understanding the nature and mitigation of software bugs in these systems has grown correspondingly. Addressing the challenges of software maintenance in autonomous driving systems (e.g., handling real-time system decisions and ensuring safety-critical reliability) is crucial due to the unique combination of real-time decision-making requirements and the high stakes of operational failures in ADSes. The potential of automated tools in this domain is promising, yet there remains a gap in our comprehension of the challenges faced and the strategies employed during manual debugging and repair of such systems. In this paper, we present an empirical study that investigates bug-fix patterns in ADSes, with the aim of improving reliability and safety. We have analyzed the commit histories and bug reports of two major autonomous driving projects, Apollo and Autoware, from 1,331 bug fixes with the study of bug symptoms, root causes, and bug-fix patterns. Our study reveals several dominant bug-fix patterns, including those related to path planning, data flow, and configuration management. Additionally, we find that the frequency distribution of bug-fix patterns varies significantly depending on their nature and types and that certain categories of bugs are recurrent and more challenging to exterminate. Based on our findings, we propose a hierarchy of ADS bugs and two taxonomies of 15 syntactic bug-fix patterns and 27 semantic bug-fix patterns that offer guidance for bug identification and resolution. We also contribute a benchmark of 1,331 ADS bug-fix instances.
2502.01940
Toward a Low-Cost Perception System in Autonomous Vehicles: A Spectrum Learning Approach
cs.CV eess.IV
We present a cost-effective new approach for generating denser depth maps for Autonomous Driving (AD) and Autonomous Vehicles (AVs) by integrating the images obtained from deep neural network (DNN) 4D radar detectors with conventional camera RGB images. Our approach introduces a novel pixel positional encoding algorithm inspired by Bartlett's spatial spectrum estimation technique. This algorithm transforms both radar depth maps and RGB images into a unified pixel image subspace called the Spatial Spectrum, facilitating effective learning based on their similarities and differences. Our method effectively leverages high-resolution camera images to train radar depth map generative models, addressing the limitations of conventional radar detectors in complex vehicular environments, thus sharpening the radar output. We develop spectrum estimation algorithms tailored for radar depth maps and RGB images, a comprehensive training framework for data-driven generative models, and a camera-radar deployment scheme for AV operation. Our results demonstrate that our approach also outperforms the state-of-the-art (SOTA) by 27.95% in terms of Unidirectional Chamfer Distance (UCD).