Se-Young Yun

Temporal Alignment Guidance: On-manifold Sampling in Diffusion Models

Predicting LLM Reasoning Performance with Small Proxy Model

Process-Verified Reinforcement Learning for Theorem Proving via Lean

Revisiting Multi-Agent Debate as Test-Time Scaling: When Does Multi-Agent Help?

Toward Self-Evolving Systems of LLM Agents through Exploration and Iterative Feedback

Gradient Fan-in Asymmetry: The Structural Cause of Layer Redundancy in Deep Transformers

KLASS: KL-Guided Fast Inference in Masked Diffusion Models

GL-LowPopArt: A Nearly Instance-Wise Minimax-Optimal Estimator for Generalized Low-Rank Trace Regression

Flex-Judge: Text-Only Reasoning Unleashes Zero-Shot Multimodal Evaluators

Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation

Revisiting Instance-Optimal Cluster Recovery in the Labeled Stochastic Block Model

DistiLLM-2: A Contrastive Approach Boosts the Distillation of LLMs

AdaSTaR: Adaptive Data Sampling for Training Self-Taught Reasoners

Multi-Task Corrupted Prediction for Learning Robust Audio-Visual Speech Representation

FlickerFusion: Intra-trajectory Domain Generalizing Multi-agent Reinforcement Learning

Efficient Parametric SVD of Koopman Operator for Stochastic Dynamical Systems

QuRe: Query-Relevant Retrieval through Hard Negative Sampling in Composed Image Retrieval

MA$^2$E: Addressing Partial Observability in Multi-Agent Reinforcement Learning with Masked Auto-Encoder

Attic: A New Architecture for Tabular In-Context Learning Transformers

Automated Filtering of Human Feedback Data for Aligning Text-to-Image Diffusion Models

Understanding Bias Reinforcement in LLM Agents Debate

MoHAVE: Mixture of Hierarchical Audio-Visual Experts for Robust Speech Recognition

DPM: Dual Preferences-based Multi-Agent Reinforcement Learning

CIRQRS: Evaluating Query Relevance Score in Composed Image Retrieval

A Unified Framework for Speculative Decoding with Multiple Drafters as a Bandit

Fine-tuned In-Context Learning Transformers are Excellent Tabular Data Classifiers.