Papers

12094 papers

ICLR2025

Differentiable Integer Linear Programming

Summary pending...

Integer Linear ProgrammingLearning to Optimize
ICLR2025

Straight to Zero: Why Linearly Decaying the Learning Rate to Zero Works Best for LLMs

Summary pending...

Learning rate schedulesLarge language models (LLMs)AdamW optimizer
ICLR2025

ActSafe: Active Exploration with Safety Constraints for Reinforcement Learning

Summary pending...

Safe ExplorationConstrained Markov Decision ProcessesSafe Reinforcement Learning
ICLR2025

Ctrl-Adapter: An Efficient and Versatile Framework for Adapting Diverse Controls to Any Diffusion Model

Summary pending...

AdapterDiffusionControlNet
ICLR2025

LevAttention: Time, Space and Streaming Efficient Algorithm for Heavy Attentions

Summary pending...

transformersattentionrandomized linear algebra
ICLR2025

Port-Hamiltonian Architectural Bias for Long-Range Propagation in Deep Graph Networks

Summary pending...

graph representation learninglong-range propagationordinary differential equations
ICLR2025

A Transfer Attack to Image Watermarks

Summary pending...

Image WatermarkingTransfer AttackAI-generated Image
ICLR2025

Air Quality Prediction with Physics-Guided Dual Neural ODEs in Open Systems

Summary pending...

Air Quality Prediction; Physics-guided Deep Learning
ICLR2025

MUSE: Machine Unlearning Six-Way Evaluation for Language Models

Summary pending...

Language ModelsMachine Unlearning
ICLR2025

Composable Interventions for Language Models

Summary pending...

Model editingCompressionUnlearning
ICLR2025

Generative Monoculture in Large Language Models

Summary pending...

monoculturebiasalignment
ICLR2025

Training Robust Ensembles Requires Rethinking Lipschitz Continuity

Summary pending...

robustnesslipschitznessensembles
ICLR2025

BiGR: Harnessing Binary Latent Codes for Image Generation and Improved Visual Representation Capabilities

Summary pending...

Image generationGenerative modelRepresentation learning
ICLR2025

Sparse Autoencoders Do Not Find Canonical Units of Analysis

Summary pending...

sparse autoencodersmechanistic interpretability
ICLR2025

Task-Adaptive Pretrained Language Models via Clustered-Importance Sampling

Summary pending...

task-adaptive pretraininglanguage modelsimportance sampling
ICLR2025

Autoregressive Pretraining with Mamba in Vision

Summary pending...

Auto regressive Pretraining
ICLR2025

Dualformer: Controllable Fast and Slow Thinking by Learning with Randomized Reasoning Traces

Summary pending...

planningreasoningSequential Decision Making
ICLR2025

Human-Aligned Chess With a Bit of Search

Summary pending...

chessalignmentadaptive MCTS
ICLR2025

The Superposition of Diffusion Models Using the Itô Density Estimator

Summary pending...

generative modellingprotein generationimage generation
ICLR2025

Topological Zigzag Spaghetti for Diffusion-based Generation and Prediction on Graphs

Summary pending...

Graph learningTopological Data AnalysisGeometric Deep Learning