Papers

12094 papers

ICML2025

Noisy SIGNSGD Is More Differentially Private Than You (Might) Think

Summary pending...

differential privacysign-based compressor
ICML2025

Surrogate Prompt Learning: Towards Efficient and Diverse Prompt Learning for Vision-Language Models

Summary pending...

transfer learningprompt learningvision-language model
ICML2025

Extreme Value Policy Optimization for Safe Reinforcement Learning

Summary pending...

constrained RLextreme value theoryconstraint satisfaction
ICML2025

PoisonBench: Assessing Language Model Vulnerability to Poisoned Preference Data

Summary pending...

backdoor attackdata poisoningpreference learning
ICML2025

A Theoretical Study of (Hyper) Self-Attention through the Lens of Interactions: Representation, Training, Generalization

Summary pending...

attentionself-attentiondeep learning
ICML2025

Feature Shift Localization Network

Summary pending...

feature shift localizationshift detectionshift
ICML2025

LAION-C: An Out-of-Distribution Benchmark for Web-Scale Vision Models

Summary pending...

OODrepresentation learningbenchmark
ICML2025

Sorbet: A Neuromorphic Hardware-Compatible Transformer-Based Spiking Language Model

Summary pending...

Spiking Neural Networks; Small Language Models; Energy Efficiency
ICML2025

Equivariant Neural Tangent Kernels

Summary pending...

neural tangent kernelequivariant neural networksgeometric deep learning
ICML2025

Survival Analysis via Density Estimation

Summary pending...

survival analysistime-to-event analysiscompeting risks
ICML2025

Maximizing Intermediate Checkpoint Value in LLM Pretraining with Bayesian Optimization

Summary pending...

Large Language Models (LLMs)PretrainingCheckpoint Merging
ICML2025

Otter: Generating Tests from Issues to Validate SWE Patches

Summary pending...

LLMsSWE PatchesReproduction Tests
ICML2025

Learning Minimum-Size BDDs: Towards Efficient Exact Algorithms

Summary pending...

parameterized complexityNP-hard problembinary decision diagrams
ICML2025

Tool Unlearning for Tool-Augmented LLMs

Summary pending...

Tool-Augmented LLMMachine Unlearning
ICML2025

Dataflow-Guided Neuro-Symbolic Language Models for Type Inference

Summary pending...

type inferenceneuro-symboliclanguage models
ICML2025

Joint Metric Space Embedding by Unbalanced Optimal Transport with Gromov–Wasserstein Marginal Penalization

Summary pending...

embedded Wasserstein distanceGromov-Wasserstein distanceoptimal transport
ICML2025

Transfer Q-Learning with Composite MDP Structures

Summary pending...

Transfer learningQ learningUCB Algorithms
ICML2025

Beyond Low-rank Decomposition: A Shortcut Approach for Efficient On-Device Learning

Summary pending...

Deep LearningComputer VisionCompression
ICML2025

Policy Optimization for CMDPs with Bandit Feedback: Learning Stochastic and Adversarial Constraints

Summary pending...

CMDPsOnline Learning
ICML2025

Robust Offline Reinforcement Learning with Linearly Structured $f$-Divergence Regularization

Summary pending...

Reinforcement LearningRobust Markov Decision Process