arXivSub Start free trial

NeurIPS 2025 Papers with Code β€” Page 2

Conference on Neural Information Processing Systems Β· 2283 papers

Adjacent Words, Divergent Intents: Jailbreaking Large Language Models via Task Concurrency

Yukun Jiang (CISPA Helmholtz Center for Information Security), Yang Zhang (CISPA Helmholtz Center for Information Security)

CodeAdversarial AttackTransformerLarge Language ModelTextBenchmark

🎯 What it does: A word-level task concurrency-based LLM jailbreak framework called JAIL-CON is proposed, demonstrating that LLMs can still answer tasks under concurrent interactions while reducing the identification rate of security protections.

Adjoint SchrΓΆdinger Bridge Sampler

Guan-Horng Liu (Meta), Ricky T. Q. Chen (Meta)

CodeGenerationOptimizationDrug DiscoveryGraph Neural NetworkDiffusion modelGraphPhysics RelatedStochastic Differential Equation

🎯 What it does: A novel Diffusion Sampler called the Adjoint Schrâdinger Bridge Sampler (ASBS) is proposed for sampling Boltzmann distributions given only the energy function.

Adjusting Initial Noise to Mitigate Memorization in Text-to-Image Diffusion Models

Hyeonggeun Han (Seoul National University), Jungwoo Lee (Seoul National University)

CodeGenerationData SynthesisDiffusion modelImageText

🎯 What it does: This paper proposes to alleviate the memorization problem in text-to-image diffusion models by adjusting the initial noise samples and presents two mitigation strategies during inference (batch-wise and per-sample).

ADPretrain: Advancing Industrial Anomaly Detection via Anomaly Representation Pretraining

Xincheng Yao (Shanghai Jiao Tong University), Chongyang Zhang (Shanghai Jiao Tong University)

CodeAnomaly DetectionTransformerContrastive LearningImage

🎯 What it does: A pre-training framework for industrial anomaly detection, ADPretrain, is proposed, utilizing large-scale industrial data RealIAD and residual features for angle and norm contrastive learning to obtain more discriminative pre-trained representations.

Adv-SSL: Adversarial Self-Supervised Representation Learning with Theoretical Guarantees

Chenguang Duan (Wuhan University), Jerry Zhijian Yang (Wuhan University)

CodeRepresentation LearningAdversarial AttackContrastive LearningImage

🎯 What it does: A new adversarial self-supervised representation learning framework, Adv-SSL, is proposed, which utilizes unbiased min-max optimization to eliminate estimation bias in traditional covariance regularization, thereby enhancing representation transfer performance.

Advancing Machine-Generated Text Detection from an Easy to Hard Supervision Perspective

Chenwang Wu (Hong Kong Baptist University), Defu Lian (University of Science and Technology of China)

CodeClassificationOptimizationLarge Language ModelSupervised Fine-TuningText

🎯 What it does: A framework for easy-to-difficult supervised enhancement is proposed, which improves the performance of machine-generated text detection by constructing long-text supervisors.

Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment

Xiaojun Jia (Nanyang Technological University), Yang Liu (Nanyang Technological University)

CodeAdversarial AttackTransformerLarge Language ModelVision Language ModelImageMultimodality

🎯 What it does: A transferable attack method for closed-source multimodal large language models, called FOA-Attack, is proposed, which generates target attack samples through optimal alignment of global and local features.

Adversarial Graph Fusion for Incomplete Multi-view Semi-supervised Learning with Tensorial Imputation

Zhangqi Jiang (National University of Defense Technology), Xinyan Liang (Shanxi University)

CodeClassificationOptimizationGraph Neural NetworkGenerative Adversarial NetworkGraph

🎯 What it does: An algorithm AGF-TI for incomplete multi-view semi-supervised learning is proposed to address the sub-cluster problem (SCP) caused by missing views, achieving more robust label propagation through graph fusion and missing information recovery.

Adversarial Paraphrasing: A Universal Attack for Humanizing AI-Generated Text

Yize CHENG, Soheil Feizi (University of Maryland)

CodeGenerationAdversarial AttackTransformerLarge Language ModelPrompt EngineeringText

🎯 What it does: This paper proposes a training-free, general adversarial paraphrasing framework that utilizes instruction-tuned LLMs to select the most 'human-like' token at each step of generation based on the scores from AI text detectors, effectively circumventing various AI text detectors.

Adversary Aware Optimization for Robust Defense

Daniel Wesego (University of Illinois Chicago), Pedram Rooshenas (University of Illinois Chicago)

CodeOptimizationAdversarial AttackDiffusion modelScore-based ModelImage

🎯 What it does: A diffusion model-based adversarial purification framework called AAOpt is proposed, which utilizes a pre-trained diffusion prior and a learned adversarial perturbation score network for MAP optimization purification during testing.

Afterburner: Reinforcement Learning Facilitates Self-Improving Code Efficiency Optimization

Mingzhe Du (Nanyang Technological University), See-Kiong Ng (National University of Singapore)

CodeOptimizationAI Code AssistantReinforcement Learning from Human FeedbackLarge Language ModelSupervised Fine-TuningReinforcement LearningText

🎯 What it does: An iterative optimization framework (Afterburner) is proposed, which continuously generates, evaluates, and provides feedback on code during inference using LLMs to improve code efficiency.

AgentAuditor: Human-level Safety and Security Evaluation for LLM Agents

Hanjun Luo (New York University), Hanan Salam (New York University)

CodeSafty and PrivacyTransformerLarge Language ModelTextBenchmarkRetrieval-Augmented GenerationChain-of-Thought

🎯 What it does: Proposes the AgentAuditor framework and ASSEBench benchmark for evaluating the safety and security of LLM agents;

AgentBreeder: Mitigating the AI Safety Risks of Multi-Agent Scaffolds via Self-Improvement

J Rosser (University of Oxford), Jakob Nicolaus Foerster

CodeOptimizationSafty and PrivacyLarge Language ModelAgentic AIText

🎯 What it does: AGENTBREEDER is proposed, utilizing multi-objective evolutionary search to generate multi-agent scaffolds and evaluate their capabilities and safety.

Agentic RL Scaling Law: Spontaneous Code Execution for Mathematical Problem Solving

Xinji Mai (Fudan University), Wenqiang Zhang (Fudan University)

CodeLarge Language ModelReinforcement LearningAgentic AIText

🎯 What it does: This study trains a foundational large language model using Zero Reward Reinforcement Learning (ZeroRL) to autonomously invoke a Python code interpreter to complete mathematical reasoning tasks.

AgentTTS: Large Language Model Agent for Test-time Compute-optimal Scaling Strategy in Complex Tasks

Fali Wang (Pennsylvania State University), Suhang Wang (Pennsylvania State University)

CodeOptimizationComputational EfficiencyHyperparameter SearchTransformerLarge Language ModelAgentic AIText

🎯 What it does: This paper presents AgentTTS, an LLM-based agent framework for optimal allocation of computational budgets in multi-stage complex tasks, thereby enhancing overall task performance.

Aggregation Hides Out-of-Distribution Generalization Failures from Spurious Correlations

Olawale Elijah Salaudeen (Massachusetts Institute of Technology), Marzyeh Ghassemi (Massachusetts Institute of Technology)

CodeDomain AdaptationImageBenchmark

🎯 What it does: This paper proposes an unsupervised subset selection method called OODSelect, which aims to discover hidden subsets in OOD data, revealing a negative correlation between ID and OOD accuracy, thereby uncovering the issue of spurious correlations masked by aggregated statistics.

AI Debate Aids Assessment of Controversial Claims

Salman Rahman (University of California), Saadia Gabriel (University of California)

CodeTransformerLarge Language ModelText

🎯 What it does: The study investigates the impact of AI Debate on biased reviewers in the fact judgment tasks related to COVID-19 and climate change controversies.

AI Research Agents for Machine Learning: Search, Exploration, and Generalization in MLE-bench

Edan Toledo (Meta), Yoram Bachrach (Meta)

CodeLarge Language ModelAgentic AITabularBenchmark

🎯 What it does: This paper proposes a framework for designing AI research agents as a combination of search strategies and operators, and systematically evaluates different combinations on MLE-Bench Lite.

AI-Generated Video Detection via Perceptual Straightening

Christian InternΓ² (Bielefeld University), David Klindt

CodeClassificationAnomaly DetectionTransformerContrastive LearningVideoBenchmark

🎯 What it does: An AI video detection method called ReStraV is proposed, which is based on the geometric curvature of visual representations. It utilizes a pre-trained DINOv2 visual encoder to extract representations of video frames, calculates curvature and distance statistics in the representation space as features to distinguish between real and fake videos.

AI-Researcher: Autonomous Scientific Innovation

Jiabin Tang (University of Hong Kong), Chao Huang (University of Hong Kong)

CodeRecommendation SystemTransformerLarge Language ModelAgentic AITextBenchmarkRetrieval-Augmented Generation

🎯 What it does: The AI-Researcher framework is proposed, achieving full-process automation from literature retrieval, idea generation, algorithm implementation to automatic paper writing, and constructing the Scientist-Bench benchmark.

AION-1: Omnimodal Foundation Model for Astronomical Sciences

Liam Holden Parker, Shirley Ho (Flatiron Institute)

CodeSegmentationGenerationRetrievalTransformerContrastive LearningImageMultimodality

🎯 What it does: AION-1 has been constructed, a large-scale multimodal foundation model capable of processing 39 different astronomical data modalities (images, spectra, annotations, numerical data, etc.), and self-supervised pre-training on 200 million celestial objects is achieved through a two-stage process.

AlignedGen: Aligning Style Across Generated Images

Jiexuan Zhang (Peking University), Jian Zhang (Peking University)

CodeGenerationData SynthesisTransformerDiffusion modelImage

🎯 What it does: A training-free, DiT-based style-aligned image generation framework called AlignedGen is proposed, enhancing the style consistency of generated images under different text prompts.

Aligning Evaluation with Clinical Priorities: Calibration, Label Shift, and Error Costs

Gerardo Flores, Ashia C. Wilson (Massachusetts Institute of Technology)

CodeClassificationOptimizationExplainability and InterpretabilityBiomedical DataElectronic Health Records

🎯 What it does: A framework is proposed that combines calibration, label shift, and cost of error to evaluate the practical utility of clinical decision support models using adjusted logarithmic loss.

Aligning Text-to-Image Diffusion Models to Human Preference by Classification

Longquan Dai (Nanjing University of Science and Technology), Jinhui Tang (Nanjing University of Science and Technology)

CodeClassificationGenerationReinforcement Learning from Human FeedbackDiffusion modelImage

🎯 What it does: Proposes to transform the text-to-image diffusion model alignment task into a classification problem, and achieves human preference alignment through the ABC (Alignment by Classification) framework;

Alignment of Large Language Models with Constrained Learning

Botong Zhang (University of Pennsylvania), Alejandro Ribeiro (University of Pennsylvania)

CodeOptimizationReinforcement Learning from Human FeedbackTransformerLarge Language ModelReinforcement LearningText

🎯 What it does: This paper proposes and implements an iterative alignment method based on Lagrangian duality (CAID) to achieve the alignment problem of maximizing primary rewards while satisfying secondary constraints in large language models.

ALINE: Joint Amortization for Bayesian Inference and Active Data Acquisition

Daolang Huang (Aalto University), Luigi Acerbi (University of Helsinki)

CodeHyperparameter SearchTransformerReinforcement LearningTabular

🎯 What it does: ALINE is proposed, a unified framework for amortized Bayesian inference and active data acquisition that allows for inference while querying data in real-time.

AliO: Output Alignment Matters in Long-Term Time Series Forecasting

Kwangryeol Park (Ulsan National Institute of Science and Technology), Seulki Lee (Ulsan National Institute of Science and Technology)

CodeTransformerTime Series

🎯 What it does: The AliO method is proposed to address the output alignment problem in long sequence forecasting, and the TAM metric is introduced to quantify alignment quality.

ALMGuard: Safety Shortcuts and Where to Find Them as Guardrails for Audio–Language Models

Weifei Jin (Beijing University of Posts and Telecommunications), Derui Wang (CSIRO Data61 Responsible AI Research Centre)

CodeSafty and PrivacyAdversarial AttackLarge Language ModelAudio

🎯 What it does: Proposes ALMGuard, which activates safety shortcuts in audio language models to protect the model from jailbreak attacks using universal audio perturbations.

AlphaBeta is not as good as you think: a simple random games model for a better analysis of deterministic game-solving algorithms

Raphael Boige, Bruno Scherrer (UniversitΓ© de Lorraine)

Code

🎯 What it does: A forward random game tree model is proposed, which constructs the tree step by step through hierarchical conditional distributions while enforcing minimax constraints, to evaluate the average complexity of deterministic game solving algorithms.

AlphaDecay: Module-wise Weight Decay for Heavy-Tailed Balancing in LLMs

Di He (Shenzhen Institute of Advanced Technology), Lu Yin (University of Surrey)

CodeTransformerLarge Language ModelText

🎯 What it does: The AlphaDecay method is proposed, which dynamically allocates different weight decay coefficients to different modules in large language models (LLMs), evaluating the spectral characteristics of each module based on the Heavy-Tail Self-Regularization (HT-SR) theory, thereby achieving module-level weight decay scheduling.

AltLoRA: Towards Better Gradient Approximation in Low-Rank Adaptation with Alternating Projections

Xin Yu (Pennsylvania State University), Lingzhou Xue (Pennsylvania State University)

CodeTransformerLarge Language ModelSupervised Fine-TuningText

🎯 What it does: This paper proposes AltLoRA, a parameter-efficient fine-tuning method that alternately updates the gradients and momentum of LoRA in a low-rank space, maintaining extremely low memory overhead.

ALTo: Adaptive-Length Tokenizer for Autoregressive Mask Generation

Lingfeng Wang (Uni-Ubi), Wuyue Zhao (Uni-Ubi)

CodeSegmentationGenerationTransformerLarge Language ModelSupervised Fine-TuningReinforcement LearningMultimodality

🎯 What it does: Proposes the ALTo adaptive length masking tokenizer and integrates it into a multimodal large language model, achieving dynamic generation of masked tokens based on target complexity;

Ambient Diffusion Omni: Training Good Models with Bad Data

Giannis Daras (Massachusetts Institute of Technology), Constantinos Costis Daskalakis

CodeGenerationData SynthesisDiffusion modelImageText

🎯 What it does: Proposes the Ambient Diffusion Omni (Ambient-o) framework, which enhances image generation quality and diversity in diffusion model training using low-quality, synthetic, or outlier distribution data.

Ambient Proteins - Training Diffusion Models on Noisy Structures

Giannis Daras (Massachusetts Institute of Technology), Daniel Jesus Diaz

CodeProtein Structure PredictionDiffusion modelMesh

🎯 What it does: This paper proposes the Ambient Protein Diffusion framework, which utilizes low-confidence AlphaFold structures as noisy training samples to train a protein diffusion model that generates new proteins with high diversity and high designability.

AmorLIP: Efficient Language-Image Pretraining via Amortization

Haotian Sun (Georgia Institute of Technology), Bo Dai (Georgia Institute of Technology)

CodeRetrievalComputational EfficiencyRepresentation LearningTransformerVision Language ModelContrastive LearningImageTextMultimodality

🎯 What it does: The AMORLIP framework is proposed, which eliminates the need for a large number of negative samples in CLIP pre-training through a lightweight amortization network, achieving more efficient language-image alignment learning.

Amortized Variational Transdimensional Inference

Laurence Davies (University of New South Wales), Scott A Sisson

CodeFlow-based ModelTabularTime Series

🎯 What it does: A CoSMIC regularization flow and VTI framework capable of adaptive variational inference in cross-dimensional (multi-model) space is proposed, which can approximate the posterior distribution of different dimensional models using a single variational density.

Amplifying Prominent Representations in Multimodal Learning via Variational Dirichlet Process

Tsai Hor Chan (University of Pennsylvania), Lequan Yu (University of Hong Kong)

CodeGenerationAnomaly DetectionRepresentation LearningConvolutional Neural NetworkRecurrent Neural NetworkTransformerMultimodalityBiomedical DataElectronic Health Records

🎯 What it does: A multimodal learning framework DPMM based on the Dirichlet process is designed, which utilizes its richer-gets-richer property to amplify significant features in multimodal feature fusion and achieve adaptive alignment of modality distributions, while also being able to perform generative imputation for missing modalities.

An Adaptive Algorithm for Bilevel Optimization on Riemannian Manifolds

Xu Shi (Fudan University), Rujun Jiang (Fudan University)

CodeOptimizationTabular

🎯 What it does: An adaptive Riemannian double-layer optimization algorithm, AdaRHD, is proposed to solve double-layer optimization problems with lower-layer strong convexity constraints on Riemannian manifolds, without the need to know parameters such as gradients and curvatures in advance.

An Analysis of Causal Effect Estimation using Outcome Invariant Data Augmentation

UZAIR AKBAR, Bo Dai (Max Planck Institute for Intelligent Systems)

CodeDomain AdaptationContrastive LearningImageTabular

🎯 What it does: This paper proposes a method that treats data augmentation (DA) as a soft intervention to estimate unidentifiable causal effects and enhance the generalization performance of external interventions.

An Efficient Local Search Approach for Polarized Community Discovery in Signed Networks

Linus Aronsson (Chalmers University of Technology & University of Gothenburg), Morteza Haghir Chehreghani (Chalmers University of Technology & University of Gothenburg)

CodeOptimizationGraph Neural NetworkGraph

🎯 What it does: A new polarization community detection (PCD) objective function is proposed, incorporating a square size penalty to avoid clustering imbalance, and the first scalable local search algorithm is designed, achieving linear convergence rate based on block coordinate Frank-Wolfe; an efficient implementation (Alg.3) is also provided, enabling rapid iterations on large-scale signed networks.

An Efficient Orlicz-Sobolev Approach for Transporting Unbalanced Measures on a Graph

Tam Le (Institute of Statistical Mathematics), Kenji Fukumizu (Institute of Statistical Mathematics)

CodeOptimizationComputational EfficiencyGraph Neural NetworkTextGraph

🎯 What it does: This paper proposes two novel transport distancesβ€”Orlicz-EPT (entropy-perturbed transport utilizing the Orlicz geometric structure) and Orlicz-Sobolev Transport (OST)β€”to address optimal transport problems for measures with different total masses (unbalanced) on graphs; it also provides an efficient algorithm for solving OST through one-dimensional optimization and proves its relationship with traditional distances such as OW, GST, ST, and UST; theoretical proofs and experimental validations demonstrate its computational efficiency and performance.

An Evidence-Based Post-Hoc Adjustment Framework for Anomaly Detection Under Data Contamination

Sukanya Patra (University of Mons), Souhaib Ben Taieb (Mohamed bin Zayed University of Artificial Intelligence)

CodeAnomaly DetectionContrastive LearningImageTabular

🎯 What it does: This work proposes a post-adjustment framework EPHAD for adaptive correction when testing unsupervised anomaly detection models under training data contamination.

An Investigation of Memorization Risk in Healthcare Foundation Models

Sana Tonekaboni (Massachusetts Institute of Technology), Marzyeh Ghassemi (Massachusetts Institute of Technology)

CodeSafty and PrivacyBiomedical DataElectronic Health Records

🎯 What it does: This study investigates the memory risks in the foundational model of structured electronic health records and proposes a black-box evaluation testing framework.

AnaCP: Toward Upper-Bound Continual Learning via Analytic Contrastive Projection

Saleh Momeni (University of Illinois Chicago), Bing Liu (University of Illinois Chicago)

CodeClassificationRepresentation LearningContrastive LearningImage

🎯 What it does: The AnaCP method is proposed, which utilizes analytical contrastive projection for continuous adaptation of features extracted from pre-trained models, achieving gradient-free training for class-incremental learning.

Analog Foundation Models

Julian BΓΌchel (IBM Research), Abu Sebastian (IBM Research)

CodeComputational EfficiencyKnowledge DistillationTransformerLarge Language ModelText

🎯 What it does: The study adapts large language models (such as Phi-3-mini-4k-instruct and Llama-3.2-1B-Instruct) to simulate memory computing (AIMC) hardware, achieving robustness against noise, quantization, and other non-idealities through hardware-aware training, and validates their performance on various benchmarks.

Analytic Energy-Guided Policy Optimization for Offline Reinforcement Learning

Jifeng Hu (Jilin University), Dacheng Tao (Nanyang Technological University)

CodeOptimizationReinforcement LearningDiffusion modelTabularBenchmark

🎯 What it does: A strategy optimization framework based on analytical energy guidance (AEPO) is proposed, which uses diffusion models in offline reinforcement learning to achieve more precise action sampling through analytical intermediate energy.

Anchor-based Maximum Discrepancy for Relative Similarity Testing

Zhijian Zhou (University of Melbourne), Feng Liu (University of Melbourne)

CodeClassificationAnomaly DetectionConvolutional Neural NetworkGenerative Adversarial NetworkImage

🎯 What it does: The anchor-based maximum discrepancy (AMD) method is proposed, which simultaneously learns the relative similarity hypothesis and the optimal kernel, addressing the issues of kernel selection and hypothesis prior in traditional relative similarity testing.

AngleRoCL: Angle-Robust Concept Learning for Physically View-Invariant Adversarial Patches

Wenjun Ji (Nankai University), Qing Guo (Nankai University)

CodeObject DetectionGenerationAdversarial AttackDiffusion modelImage

🎯 What it does: The AngleRoCL method is proposed, achieving high attack effectiveness of text-to-image generation attack patches under different perspectives.

Angles Don’t Lie: Unlocking Training‑Efficient RL Through the Model’s Own Signals

Qinsi Wang (Duke University), Yiran Chen (Duke University)

CodeLarge Language ModelReinforcement LearningTabularSequential

🎯 What it does: Proposed a GAIN-RL framework based on the model's own perspective of signal concentration for dynamic scheduling of reinforcement learning fine-tuning data.

Angular Steering: Behavior Control via Rotation in Activation Space

Hieu M. Vu (Torilab), Tan Minh Nguyen

CodeOptimizationExplainability and InterpretabilityTransformerLarge Language ModelText

🎯 What it does: This paper proposes Angular Steering, a method for fine-grained, continuous adjustment of large language model behavior by rotating activation vectors within a two-dimensional subspace, and presents an adaptive variant.

AnimateQR: Bridging Aesthetics and Functionality in Dynamic QR Code Generation

Guangyang Wu (Shanghai Jiao Tong University), Xiaohong Liu (Shanghai Jiao Tong University)

CodeGenerationDiffusion modelImage

🎯 What it does: The AnimateQR framework is proposed, capable of generating dynamic QR codes that balance aesthetics and readability.

Anomaly Detection by an Ensemble of Random Pairs of Hyperspheres

Walid Durani (LMU Munich), Christian BΓΆhm (University of Vienna)

CodeAnomaly DetectionTabular

🎯 What it does: This paper proposes an isolation-based anomaly detection method for hypersphere sets called ADERH, which is based on random pairing.

Any Large Language Model Can Be a Reliable Judge: Debiasing with a Reasoning-based Bias Detector

Haoyan Yang (New York University), Taha Kass-Hout (GE Healthcare)

CodeTransformerLarge Language ModelSupervised Fine-TuningText

🎯 What it does: This paper proposes an external module called the Reasoning-based Bias Detector (RBD) to detect biases in LLM evaluations and provide structured reasoning to help evaluators self-correct.

AOR: Anatomical Ontology-Guided Reasoning for Medical Large Multimodal Model in Chest X-Ray Interpretation

Qingqiu Li (Fudan University), Shujun Wang (Hong Kong Polytechnic University)

CodeRecognitionObject DetectionSegmentationTransformerLarge Language ModelVision Language ModelImageTextMultimodalityBiomedical DataChain-of-Thought

🎯 What it does: A multi-step reasoning framework AOR based on anatomical ontology is proposed, supporting region-level prompts for medical multimodal large models for chest X-ray interpretation.

APML: Adaptive Probabilistic Matching Loss for Robust 3D Point Cloud Reconstruction

Sasan Sharifipour (University of Oulu), Miguel Bordallo Lopez

CodeGenerationData SynthesisPoint Cloud

🎯 What it does: This paper proposes a differentiable, near-quadratic complexity Adaptive Probability Matching Loss (APML) to replace the traditional Chamfer distance for supervised 3D point cloud reconstruction and generation tasks.

APOLLO: Automated LLM and Lean Collaboration for Advanced Formal Reasoning

Azim Ospanov (Huawei Hong Kong Research Center), Roozbeh Yousefzadeh (Chinese University of Hong Kong)

CodeLarge Language ModelTextBenchmark

🎯 What it does: This paper proposes a fully automated framework called Apollo, which utilizes LLMs, the Lean compiler, and automated solvers to collaboratively repair and generate formal proofs.

Approximately Aligned Decoding

Daniel Melcer (Northeastern University), Anoop Deoras (Amazon)

CodeGenerationComputational EfficiencyTransformerLarge Language ModelReinforcement LearningText

🎯 What it does: A new generation method called Approximately Aligned Decoding (AprAD) is proposed to avoid errors or constraint violations during the generation process of large language models while reducing computational overhead.

Approximating Shapley Explanations in Reinforcement Learning

Daniel Beechey (University of Bath), Γ–zgΓΌr Şimşek (University of Bath)

CodeExplainability and InterpretabilityReinforcement Learning

🎯 What it does: This paper proposes FastSVERL, a scalable parametric method for approximating Shapley values in reinforcement learning to explain the behavior, outcomes, and predictions of agents.

AR-RAG: Autoregressive Retrieval Augmentation for Image Generation

Jingyuan Qi (Virginia Tech), Lifu Huang (UC Davis)

CodeGenerationRetrievalTransformerImageMultimodalityRetrieval-Augmented Generation

🎯 What it does: A self-regressive retrieval-enhanced framework AR-RAG is proposed and implemented, which dynamically retrieves nearest neighbor visual information in units of image blocks during the generation process.

ArchCAD-400K: A Large-Scale CAD drawings Dataset and New Baseline for Panoptic Symbol Spotting

Ruifeng Luo (Arcplus East China Architectural Design and Research Institute Co., Ltd.), Xianzhong Zhao (Tongji University)

CodeObject DetectionSegmentationTransformerImageMultimodalityPoint CloudBenchmark

🎯 What it does: A large-scale CAD drawing dataset ArchCAD-400k was constructed, an efficient automatic annotation pipeline was designed, and a Dual Path Symbol Localization Framework (DPSS) was proposed for panoramic symbol localization.

Are Greedy Task Orderings Better Than Random in Continual Linear Regression?

Matan Tsipory (Technion), Daniel Soudry (Technion)

CodeOptimizationImageTabular

🎯 What it does: This paper studies the impact of task order on learning effectiveness in continuous linear regression, particularly by arranging task order through greedy strategies (maximum distance, maximum residual) and comparing it with random order.

Are Language Models Efficient Reasoners? A Perspective from Logic Programming

Andreas Opedal (ETH ZΓΌrich), Bernhard SchΓΆlkopf (MPI for Intelligent Systems)

CodeLarge Language ModelPrompt EngineeringText

🎯 What it does: Evaluate the efficiency of large language models in reasoning tasks from a logical programming perspective, mapping natural language proofs to the shortest logical reasoning paths.

Are Pixel-Wise Metrics Reliable for Computerized Tomography Reconstruction?

Tianyu Lin (Johns Hopkins University), Zongwei Zhou (Johns Hopkins University)

CodeRestorationSegmentationDiffusion modelImageBiomedical DataComputed Tomography

🎯 What it does: Proposed an evaluation metric for sparse-angle CT reconstruction based on anatomical segmentation and a corresponding supplementary framework called CARE, which can significantly enhance the integrity of anatomical structures;

AREAL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning

Wei Fu (Tsinghua University), Yi Wu (Ant Group)

CodeLarge Language ModelReinforcement LearningTextBenchmark

🎯 What it does: An entirely asynchronous RL system called AREAL is proposed for inference training of large-scale language models, completely decoupling the generation and training phases.

ARECHO: Autoregressive Evaluation via Chain-Based Hypothesis Optimization for Speech Multi-Metric Estimation

Jiatong Shi (Carnegie Mellon University), Shinji Watanabe (Carnegie Mellon University)

CodeTransformerAudio

🎯 What it does: This paper researches and implements a multi-indicator speech evaluation framework called ARECHO, which unifies various scales and types of evaluation indicators by tokenizing them and modeling them in a dynamic classification chain, supporting flexible reasoning for any subset of indicators.

ASGO: Adaptive Structured Gradient Optimization

Kang An (Rice University), Tong Zhang (University of Illinois Urbana-Champaign)

CodeOptimizationTransformerLarge Language ModelSupervised Fine-TuningText

🎯 What it does: An adaptive structure gradient optimization algorithm ASGO is proposed, along with theoretical convergence analysis and empirical validation.

Assessing the quality of denoising diffusion models in Wasserstein distance: noisy score and optimal bounds

Vahan Arsenyan (ENSAE Paris), Arnak S. Dalalyan (ENSAE Paris)

CodeRestorationGenerationData SynthesisDiffusion modelImageStochastic Differential Equation

🎯 What it does: Theoretical analysis of the quality of Denoising Diffusion Probabilistic Models (DDPM) under the Wasserstein-2 distance, providing an optimal error upper bound.

Association-Focused Path Aggregation for Graph Fraud Detection

Tian Qiu (Zhejiang University), Yang Gao (Zhejiang University)

CodeAnomaly DetectionGraph Neural NetworkGraphFinance Related

🎯 What it does: This paper studies the problem of fraud detection in graph structures and proposes a novel fraud detection framework based on Graph Path Aggregation (GPA).

Asymmetric Dual Self-Distillation for 3D Self-Supervised Representation Learning

Remco F. Leijenaar (University of Groningen), Hamidreza Kasaei (University of Groningen)

CodeKnowledge DistillationRepresentation LearningTransformerContrastive LearningPoint Cloud

🎯 What it does: Proposes the AsymDSD framework, unifying mask point modeling and invariance learning into 3D point cloud self-supervised learning.

Asymptotics of SGD in Sequence-Single Index Models and Single-Layer Attention Networks

Luca Arnaboldi (EPFL), Lenka Zdeborova

CodeOptimizationTransformerSequentialOrdinary Differential Equation

🎯 What it does: This study investigates the dynamics of the Sequence Single Index model (SSI) and the stochastic gradient descent (SGD) of single-layer attention networks, deriving a closed-form expression for population loss and analyzing the high-dimensional learning phase.

ATLAS: Autoformalizing Theorems through Lifting, Augmentation, and Synthesis of Data

Xiaoyang Liu (Shanghai Jiao Tong University), Tao Luo (Shanghai Jiao Tong University)

CodeData SynthesisKnowledge DistillationTransformerLarge Language ModelText

🎯 What it does: The ATLAS framework is proposed, which automatically generates large-scale high-quality natural language-formal language pairs of mathematical theorems and implements automated translation on Lean 4;

Attack by Yourself: Effective and Unnoticeable Multi-Category Graph Backdoor Attacks with Subgraph Triggers Pool

Jiangtong Li (Tongji University), Changjun Jiang (Tongji University)

CodeOptimizationAdversarial AttackGraph Neural NetworkGraph

🎯 What it does: A multi-class graph neural network backdoor attack method based on a subgraph trigger pool, EUMC, is proposed to achieve controllable and low-detectability attacks on node classification models.

Attention! Your Vision Language Model Could Be Maliciously Manipulated

Xiaosen Wang (Huazhong University of Science and Technology), Shudong Zhang (Xidian University)

CodeAdversarial AttackVision Language ModelImageText

🎯 What it does: Proposed and implemented a precise manipulation attack on visual-language models (VMA), which precisely controls the output sequence of VLMs by applying invisible perturbations to images.

AttentionPredictor: Temporal Patterns Matter for KV Cache Compression

Qingyue Yang (University of Science and Technology of China), Bin Li (University of Science and Technology of China)

CodeCompressionOptimizationConvolutional Neural NetworkLarge Language ModelText

🎯 What it does: Proposes AttentionPredictor, which predicts the next attention score through a learned spatiotemporal convolution model to achieve KV cache compression while maintaining LLM performance.

Attractive Metadata Attack: Inducing LLM Agents to Invoke Malicious Tools

Kanghua Mo (Guangzhou University), Zhihao li

CodeSafty and PrivacyAdversarial AttackTransformerLarge Language ModelReinforcement LearningPrompt EngineeringTextFinance Related

🎯 What it does: This paper proposes and implements an attack method (AMA) that induces agents to call malicious tools by manipulating the metadata (name, description, parameter schema) of LLM agent tools.

AudSemThinker: Enhancing Audio-Language Models Through Reasoning over Semantics of Sound

Gijs Wijngaard (Maastricht University), Michel Dumontier (Maastricht University)

CodeTransformerLarge Language ModelSupervised Fine-TuningReinforcement LearningVideoTextMultimodalityAudio

🎯 What it does: A structured reasoning phase audio-language model AUDSEMTHINKER has been constructed, and a new semantic description dataset AUDSEM has been released.

Auto-Search and Refinement: An Automated Framework for Gender Bias Mitigation in Large Language Models

Yue Xu (ShanghaiTech University), Wenjie Wang (ShanghaiTech University)

CodeGenerationOptimizationTransformerLarge Language ModelText

🎯 What it does: An automated, model-agnostic framework named FaIRMaker is proposed to generate 'Fairwords' and transform them into interpretable instructions through a sequence-to-sequence model, thereby mitigating gender bias in large language models while maintaining the original model's task performance.

AutoDiscovery: Open-ended Scientific Discovery via Bayesian Surprise

Dhruv Agarwal (University of Massachusetts Amherst), Peter Clark (Allen Institute for AI)

CodeLarge Language ModelReinforcement LearningTabular

🎯 What it does: This paper presents AUTODISCOVERY, a target-free open-ended automated scientific discovery method based on Bayesian surprise.

AutoEdit: Automatic Hyperparameter Tuning for Image Editing

Chau Pham (University at Buffalo), David Doermann (University at Buffalo)

CodeGenerationOptimizationHyperparameter SearchReinforcement LearningDiffusion modelImageBenchmark

🎯 What it does: This paper proposes AutoEdit, a reinforcement learning-based framework that automatically searches for and dynamically adjusts the best hyperparameters during the image editing process of diffusion models, avoiding the high costs of manual tuning and brute-force searching.

Autoencoding Random Forests

Binh Duc Vu (King's College London), David Watson

CodeCompressionRepresentation LearningAuto EncoderTabularBiomedical Data

🎯 What it does: This paper presents a complete method for implementing autoencoders using Random Forests (RF), including learning low-dimensional embeddings from RF and various decoding strategies;

Automatic Auxiliary Task Selection and Adaptive Weighting Boost Molecular Property Prediction

Zhiqiang Zhong (University of Luxembourg), Davide Mottin (Aarhus University)

CodeDrug DiscoveryGraph Neural NetworkTransformerLarge Language ModelPrompt EngineeringGraphBenchmark

🎯 What it does: Proposes the AUTAUT framework, which uses LLM to automatically retrieve and filter auxiliary tasks, while dynamically integrating them into the main molecular property prediction model through a gradient-aligned adaptive weighting mechanism;

Automatic Synthetic Data and Fine-grained Adaptive Feature Alignment for Composed Person Retrieval

Delong Liu (Beijing University of Posts and Telecommunications), Yuan Dong (Beijing University of Posts and Telecommunications)

CodeData SynthesisRetrievalTransformerLarge Language ModelDiffusion modelImageTextMultimodality

🎯 What it does: A new task for person retrieval combining visual and textual information is proposedβ€”Composed Person Retrieval, along with the construction of a large-scale synthetic dataset SynCPR and a manually annotated test set ITCPR.

Availability-aware Sensor Fusion via Unified Canonical Space

Dong-Hee Paek (KAIST), Seung-Hyun Kong (KAIST)

CodeObject DetectionAutonomous DrivingTransformerPoint Cloud

🎯 What it does: A usability-aware fusion method (ASF) based on unified norm projection and cross-sensor patch attention is proposed, achieving collaborative perception among cameras, LiDAR, and 4D Radar, while maintaining robustness in the event of sensor failure or degradation.

AVCD: Mitigating Hallucinations in Audio-Visual Large Language Models through Contrastive Decoding

Chaeyoung Jung (Korea Advanced Institute of Science and Technology), Joon Son Chung (Korea Advanced Institute of Science and Technology)

CodeOptimizationComputational EfficiencyTransformerLarge Language ModelContrastive LearningVideoTextMultimodalityAudio

🎯 What it does: A training-free and human intervention-free inference decoding framework called AVCD is proposed to alleviate the hallucination problem in audio-video large language models.

Axial Neural Networks for Dimension-Free Foundation Models

Hyunsu Kim (Korea Advanced Institute of Science and Technology), Juho Lee (Korea Advanced Institute of Science and Technology)

CodeGraph Neural NetworkTransformerTime SeriesBenchmarkPhysics RelatedOrdinary Differential Equation

🎯 What it does: A dimension-independent Axial Neural Network (XNN) is proposed in the context of multidimensional PDE solutions, enabling the model to be uniformly trained and inferred across different spatial dimensions such as 1D, 2D, and 3D.

Backdoor Cleaning without External Guidance in MLLM Fine-tuning

Xuankun Rong (Wuhan University), Mang Ye (Nanyang Technological University)

CodeTransformerLarge Language ModelSupervised Fine-TuningVision Language ModelMultimodality

🎯 What it does: In the fine-tuning process of multimodal large language models, the BYE framework is proposed, which utilizes the self-supervised signal of attention entropy to perform unsupervised filtering on the training set, automatically detecting and removing samples carrying backdoor triggers, thereby enhancing the model's security in attack scenarios.

Backdoor Mitigation via Invertible Pruning Masks

Kealan Dunnett (Queensland University of Technology), Raja Jurdak (Queensland University of Technology)

CodeOptimizationAdversarial AttackTransformerImage

🎯 What it does: A model pruning method based on reversible masks and component selection (IMS) is proposed to eliminate backdoor attacks in deep learning models with only a small number of clean samples.

BADiff: Bandwidth Adaptive Diffusion Model

Xi Zhang (Nanyang Technological University), Weisi Lin (Nanyang Technological University)

CodeGenerationCompressionDiffusion modelImage

🎯 What it does: This paper presents BADiff, a diffusion model that can adaptively generate image quality based on real-time network bandwidth.

BadVLA: Towards Backdoor Attacks on Vision-Language-Action Models via Objective-Decoupled Optimization

Xueyang Zhou (Huazhong University of Science and Technology), Lichao Sun (Lehigh University)

CodeOptimizationAdversarial AttackVision-Language-Action ModelContrastive LearningMultimodality

🎯 What it does: This paper studies backdoor attacks on the Vision-Language-Action (VLA) model and proposes a target decoupling two-stage optimization framework named BadVLA.

Balanced Active Inference

Boyu Chen (Xiamen University), Zhonglei Wang (Xiamen University)

CodeTabular

🎯 What it does: A balanced active inference framework is proposed, combining cube balanced sampling with active inference to achieve higher statistical efficiency by balancing model uncertainty under a limited labeling budget.

Balanced Token Pruning: Accelerating Vision Language Models Beyond Local Optimization

kaiyuan Li, Xinlei Chen (Tsinghua University)

CodeCompressionOptimizationComputational EfficiencyTransformerVision Language ModelMultimodality

🎯 What it does: A balanced visual token pruning method BTP is proposed, which can significantly compress the number of visual tokens while maintaining the performance of the visual language model.

Balancing Multimodal Training Through Game-Theoretic Regularization

Konstantinos Kontras (KU Leuven), Maarten De Vos (KU Leuven)

CodeOptimizationContrastive LearningMultimodality

🎯 What it does: A multi-modal competitive regularization method MCR is proposed, which utilizes information theory to decompose unique and shared information and dynamically balances multi-modal contributions through game theory.

BAM-ICL: Causal Hijacking In-Context Learning with Budgeted Adversarial Manipulation

Rui Chu (Tufts University), Yingjie Lao (Tufts University)

CodeClassificationAdversarial AttackTransformerLarge Language ModelText

🎯 What it does: This paper proposes a budget allocation-based two-stage attack framework (BAM-ICL), which hijacks model output by finely perturbing context examples during few-shot learning (ICL) in large language model inference.

Bandit Guided Submodular Curriculum for Adaptive Subset Selection

Prateek Chanda (Indian Institute of Technology Bombay), Ganesh Ramakrishnan (Indian Institute of Technology Bombay)

CodeOptimizationReinforcement LearningImageText

🎯 What it does: An online submodular subset selection framework called ONLINESUBMOD is proposed, enabling adaptive curriculum learning based on multi-armed bandits.

Bayesian Ego-graph inference for Networked Multi-Agent Reinforcement Learning

Wei Duan (Australian Artificial Intelligence Institute University of Technology Sydney), Junyu Xuan (Australian Artificial Intelligence Institute University of Technology Sydney)

CodeGraph Neural NetworkReinforcement LearningGraph

🎯 What it does: A decentralized actor-critic framework named BayesG is proposed, which utilizes Bayesian variational inference to learn local interaction structures, allowing each agent to sample sparse subgraphs on its ego-graph and make decisions.

Bayesian Optimization with Preference Exploration using a Monotonic Neural Network Ensemble

Hanyang Wang (University of Warwick), Matthias Poloczek (Amazon)

CodeOptimizationContrastive Learning

🎯 What it does: This study combines Bayesian optimization with preference exploration to address multi-objective black-box optimization problems, learning the decision maker's utility preferences and quickly locating the optimal solution.

Belief-Calibrated Multi-Agent Consensus Seeking for Complex NLP Tasks

Wentao Deng (Shandong University), Pengjie Ren (Shandong University)

CodeLarge Language ModelAgentic AIText

🎯 What it does: A belief calibration-based multi-agent consensus seeking framework (BCCS) is proposed, achieving stable consensus in multi-agent systems through belief calibration consensus judgment, collaborator allocation, and leader selection.

BeliefMapNav: 3D Voxel-Based Belief Map for Zero-Shot Object Navigation

Zibo Zhou (Shanghai Jiao Tong University), Siheng Chen (Shanghai Jiao Tong University)

CodeOptimizationRobotic IntelligenceTransformerLarge Language ModelVision Language ModelSimultaneous Localization and MappingPoint Cloud

🎯 What it does: This paper presents BeliefMapNav, a belief map system based on 3D voxels for zero-shot object navigation.

Benford’s Curse: Tracing Digit Bias to Numerical Hallucination in LLMs

Jiandong Shao (Nanyang Technological University), Jianfei Yang (Nanyang Technological University)

CodeLarge Language ModelTextBenchmark

🎯 What it does: This paper systematically analyzes the numerical bias exhibited by large language models in numerical reasoning tasks by constructing a unified benchmark for digital distribution. It demonstrates that the digit distribution similar to Benford's law in the pre-training corpus is internalized by the model, leading to numerical hallucinations.

Better Estimation of the Kullback--Leibler Divergence Between Language Models

Afra Amini (ETH Zurich), Ryan Cotterell (ETH Zurich)

CodeReinforcement Learning from Human FeedbackTransformerLarge Language ModelReinforcement LearningText

🎯 What it does: A Rao-Blackwellized KL divergence estimator is proposed and implemented to assess the differences between language models.