Papers
12094 papers
ICLR2025
Revisiting Source-Free Domain Adaptation: a New Perspective via Uncertainty Control
Summary pending...
Source-Free Domain AdaptationUnsupervised Domain Adaptation
ICLR2025
NetFormer: An interpretable model for recovering dynamical connectivity in neuronal population dynamics
Summary pending...
neuronal dynamicsdynamical connectivityinterpretability
ICLR2025
Block Verification Accelerates Speculative Decoding
Summary pending...
llm efficiencyspeculative decodingdistribution coupling
ICLR2025
BAMDP Shaping: a Unified Framework for Intrinsic Motivation and Reward Shaping
Summary pending...
Reinforcement Learning TheoryBayesian Reinforcement LearningIntrinsic Motivation
ICLR2025
Fine-Tuning Discrete Diffusion Models via Reward Optimization with Applications to DNA and Protein Design
Summary pending...
Discrete Diffusion ModelsReward OptimizationFine-Tuning
ICLR2025
EC-DIT: Scaling Diffusion Transformers with Adaptive Expert-Choice Routing
Summary pending...
Diffusion transformertext-to-image synthesisMixture-of-Experts
ICLR2025
Self-Improving Robust Preference Optimization
Summary pending...
Preference optimizationdirect alignmentreinforcement learning from human feedback
ICLR2025
Improved Diffusion-based Generative Model with Better Adversarial Robustness
Summary pending...
Generative Model; Adversarial Robustness; Diffusion Model; Distributional Robustness Optimization
ICLR2025
PWM: Policy Learning with Multi-Task World Models
Summary pending...
reinforcement learningmodel-based reinforcement learningcontinuous control
ICLR2025
The Semantic Hub Hypothesis: Language Models Share Semantic Representations Across Languages and Modalities
Summary pending...
interpretabilityrepresentationmultilingual
ICLR2025
AssembleFlow: Rigid Flow Matching with Inertial Frames for Molecular Assembly
Summary pending...
rigid flow matchinginertial framequaternion representation
ICLR2025
Self-MoE: Towards Compositional Large Language Models with Self-Specialized Experts
Summary pending...
Efficient Specialization of LLMsSelf-ImprovingMixture of Experts
ICLR2025
Looking Inward: Language Models Can Learn About Themselves by Introspection
Summary pending...
IntrospectionLarge Language ModelsModel awareness
ICLR2025
AVHBench: A Cross-Modal Hallucination Benchmark for Audio-Visual Large Language Models
Summary pending...
Multi-modal Large Language ModelsHallucination in Large Language ModelsAudio-visual Learning
ICLR2025
Confidence Elicitation: A New Attack Vector for Large Language Models
Summary pending...
adversarial attackadversarial robustnessconfidence elicitation.
ICLR2025
MetaMetrics: Calibrating Metrics for Generation Tasks Using Human Preferences
Summary pending...
metricshuman preferencescalibrating
ICLR2025
Contrastive Learning from Synthetic Audio Doppelgängers
Summary pending...
synthetic dataaudiocontrastive learning
ICLR2025
Multiple Heads are Better than One: Mixture of Modality Knowledge Experts for Entity Representation Learning
Summary pending...
Multi-modal Information FusionKnowledge GraphMulti-modal Entity Representation
ICLR2025
Weighted Multi-Prompt Learning with Description-free Large Language Model Distillation
Summary pending...
Prompt learningVision-language modelsLarge language models