Papers

12094 papers

ICML2025

Online Linear Classification with Massart Noise

Summary pending...

Online LearningMassart noise
ICML2025

MITIGATING OVER-EXPLORATION IN LATENT SPACE OPTIMIZATION USING LES

Summary pending...

VAEDrug-Design
ICML2025

Mastering Massive Multi-Task Reinforcement Learning via Mixture-of-Expert Decision Transformer

Summary pending...

multi-task RLoffline RLDecision Transformer
ICML2025

Schwarz–Schur Involution: Lightspeed Differentiable Sparse Linear Solvers

Summary pending...

differentiable linear solversparse solvergeneralized deconvolution
ICML2025

InfoSAM: Fine-Tuning the Segment Anything Model from An Information-Theoretic Perspective

Summary pending...

Segment Anything ModelParameter-efficient fine-tuningMutual information
ICML2025

A Sample Efficient Conditional Independence Test in the Presence of Discretization

Summary pending...

Conditional independence testdiscretizationGeneralized method of Moments
ICML2025

Iterative Vectors: In-Context Gradient Steering without Backpropagation

Summary pending...

In-Context LearningActivation SteeringBackpropagation-Free
ICML2025

Provable Maximum Entropy Manifold Exploration via Diffusion Models

Summary pending...

diffusion modelsexplorationfine-tuning
ICML2025

Progressive Tempering Sampler with Diffusion

Summary pending...

parallel temperingneural samplerssampling from unnormalised densities
ICML2025

Understanding Input Selectivity in Mamba: Impact on Approximation Power, Memorization, and Associative Recall Capacity

Summary pending...

MambaState Space ModelInput selectivity
ICML2025

When, Where and Why to Average Weights?

Summary pending...

Weight AveragingCheckpoint AveragingOptimization
ICML2025

Fast and Provable Algorithms for Sparse PCA with Improved Sample Complexity

Summary pending...

Sparse PCAprincipal component analysissingle-spiked covariance model
ICML2025

Guarantees of a Preconditioned Subgradient Algorithm for Overparameterized Asymmetric Low-rank Matrix Recovery

Summary pending...

Matrix factorisationlow-rankmatrix sensing
ICML2025

The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence

Summary pending...

Refusal directionsLarge Language ModelActivation Steering
ICML2025

MathConstruct: Challenging LLM Reasoning with Constructive Proofs

Summary pending...

LLMmathevaluation
ICML2025

The Hidden Life of Tokens: Reducing Hallucination of Large Vision-Language Models Via Visual Information Steering

Summary pending...

Large Vision-Language ModelHallucination
ICML2025

LIVS: A Pluralistic Alignment Dataset for Inclusive Public Spaces

Summary pending...

Pluralistic AlignmentText-to-Image DiffusionIntersectionality
ICML2025

Square$\chi$PO: Differentially Private and Robust $\chi^2$-Preference Optimization in Offline Direct Alignment

Summary pending...

Offline alignmentgeneral function approximationsdifferential privacy
ICML2025

Towards Theoretical Understanding of Sequential Decision Making with Preference Feedback

Summary pending...

preference-basedsequential decision makingpreference feedback
ICML2025

LAuReL: Learned Augmented Residual Layer

Summary pending...

residual connectionskip connectionresidual layer