Recent Talks 

20260819

이해영

HAG: Hierarchical Demographic Tree-based Agent Generation for Topic-Adaptive Simulation

이경선

AWS – Data Clean Room

20260813

안영민

(NeurIPS 2024) Sigmoid Gating is More Sample Efficient than Softmax Gating in Mixture of Experts 

한상문

(ICML 2025) Mixture of Experts Provably Detect and Learn the Latent Cluster Structure in Gradient-Based Learning 

박세현

(ICLR 2026) DirMoE: Dirichlet-routed Mixture of Experts

권혁규

Nonparametric estimation via diffusion models

안영민

NVIDIA – Synthetic Population

20260730

정유립

Enterprise Data Analytics Agents

김초은

Image Tokenizer

20260716
ICML Papers

이경선

TabSwift: An Efficient Tabular Foundation Model with Row-Wise Attention

정유립

Mixture of Concept Bottleneck Experts

정한교

How Can Embedding Models Bind Concepts?

김동윤

Richer Bayesian Last Layers with Subsampled NTK Features

박세현

Skip a Layer or Loop It: Learning Program of Layers in LLMs

안영민

Compositional Generalization Requires Linear Orthogonal Representations in Vision Embedding Models

이해영

The Illusion of Generalization in Tabular Language Models

김찬우

DGS-Net: Distillation-Guided Gradient Surgery for CLIP Fine-Tuning in AI-Generated Image Detection

김초은

Learning Sparse Visual Representations via Spatial-Semantic Factorization

20260630

이경선

Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction

한상문

Inlier Memorization Effect

20260605 Workshop

박세현

Energy Efficient in Mixture of Experts

한상문

Switch Transformer

이경선

LLM Acceleration

이관호

TurboQuant

이해영

Orchestrate Agent & The Agent Company

신윤섭

C-LoRA: Contextual Low-Rank Adaptation for Uncertainty Estimation in Large Language Models

Simple Factuality Probes Detect Hallucinations in Long-Form Natural Language Generation

정유립

Belief-Calibrated Multi-Agent Consensus Seeking for Complex NLP Tasks

안영민

Uncertainty as Feature Gaps

김동윤

Post-Hoc Probabilistic Vision-Language Models

정한교

Statistics in LLM

김초은

Asymptotic of Heaviside NN

diffusion DPS

김초은 & 이관호

Super Resolution

Archive by Semester