Mini Logo
AIDAS Lab Logo

AIDAS Seminar

February 13, 2026

Overview

Ongoing weekly reading seminars and technical presentations by AIDAS members.

Sessions

2026 Spring (26-1)

DateCategoryTopicPresenter
Feb 27Data SystemsvLLM-Omni: Fully Disaggregated Serving for Any-to-Any Multimodal ModelsDogeun Kim
TBDTBDGeon Choi
Feb 13RoboticsLatent Action Pretraining from VideosHoeun Lee
LLMsIntroduction to Linear AttentionYejoon Lee
Feb 6Agent SystemsLatent Collaboration in Multi-Agent SystemsSieun Hyeon
RoboticsUniversal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild RobotsHyeonggeun Kim
Jan 30ReasoningBeyond Autoregression: Discrete Diffusion for Complex Reasoning and PlanningWoohyeon Park
RoboticsEnhancing intention prediction and interpretability in service robots with LLM and KGMinwoo Kim
Jan 23LLMsProgressive Trajectory Matching for Dataset DistillationSteve Cho
ReasoningReasoning Models Generate Societies of ThoughtYongha Lee
Jan 16LLMsK-Foundation ModelsJusang Oh
LLMsGated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-FreeJinhyeok Kim
Jan 9Knowledge GraphsOntoTune: Ontology-Driven Self-training for Aligning LLMsMinteak Lim
LLMsAI models collapse when trained on recursively generated dataJaeik Kim
Jan 2LLMsArtificial Hivemind: The Open-Ended Homogeneity of Language Models (and Beyond)Woojin Kim
LLMsA Path Towards Autonomous Machine IntelligenceJihwan Hong

2025 Fall (25-2)

DateCategoryTopicPresenter
Dec 19LLMsNumerical mismatch in LLM RLYunseok Han
Diffusion ModelsTiDAR: Think in Diffusion, Talk in AutoregressionHanjun Lee
Dec 12ReasoningReward Reasoning ModelHoeun Lee
Data SystemsLightRAG: Simple and Fast Retrieval-Augmented GenerationGeon Choi
Data SystemsHiFC: High-efficiency Flash-based KV Cache Swapping for Scaling LLM InferenceDogeun Kim
Dec 5ReasoningPrismatic Synthesis: Gradient-based Data Diversification Boosts Generalization in LLM ReasoningYejoon Lee
Data SystemsHiFC: High-efficiency Flash-based KV Cache Swapping for Scaling LLM InferenceDogeun Kim
Nov 28LLMsMoBE: Mixture-of-Basis-Experts for Compressing MoE-based LLMsSieun Hyeon
RoboticsRobot Learning: A TutorialHyeonggeun Kim
Nov 21Reinforcement LearningMastering the game of Go with deep neural networks and tree searchWoohyeon Park
Nov 14ReasoningThinkMorph: Emergent Properties in Multimodal Interleaved Chain-of-Thought ReasoningYongha Lee
Knowledge GraphsA Survey on Knowledge Graphs: Representation, Acquisition, and ApplicationsMinwoo Kim
Nov 7Diffusion ModelsSparseD: Sparse Attention for Diffusion Language ModelsJinhyeok Kim
Oct 31Representation LearningConcerto: Joint 2D-3D Self-Supervised Learning Emerges Spatial RepresentationsJusang Oh
LLMsEmergent Introspective Awareness in Large Language ModelsSteve Cho
Oct 24RoboticsOn the Theoretical Limitations of Embedding-Based RetrievalMintaek Lim
Multimodal AIDeepSeek-OCR: Contexts Optical CompressionJaeik Kim
Oct 17Multimodal AIVideo models are zero-shot learners and reasonersJihwan Hong
Data SystemsPersona Vectors: Monitoring and Controlling Character Traits in Language ModelsWoojin Kim
Oct 10LLMsTraining and Inference on Any-Order Autoregressive Models the Right WayYunseok Han
Agent SystemsDynamic Planning for LLM-based Graphical User Interface AutomationHanjun Lee
Sep 26RoboticsLLaDA-VLA: Vision-Language Diffusion Action ModelsHoeun Lee
AI SafetyLLM Privacy SurveyGeun Choi
Sep 19LLMsMixture of Lookup ExpertsSieun Hyeon
LLMsSelf-Questioning Language ModelsYejoon Lee
Sep 12Representation LearningRoFormer: Enhanced Transformer with Rotary Position EmbeddingDogeun Kim
RoboticsLatent Diffusion Planning for Imitation LearningHyeonggeun Kim
Aug 22Agent SystemsAdvancing Mobile GUI Agents: A Verifier-Driven Approach to Practical DeploymentHanjun Lee
Multimodal AIUni-MoE: Scaling Unified Multimodal LLMs with Mixture of ExpertsWoohyeon Park
Aug 8LLMsSparseLoRA: Accelerating LLM Fine-Tuning with Contextual SparsityJinhyeok Kim
LLMsSubliminal Learning: Language Models Transmit Behavioral Traits via Hidden Signals in DataSteve Cho
Jul 25ReasoningTrain for the Worst, Plan for the Best: Understanding Token Ordering in Masked DiffusionsJaeik Kim
LLMsMixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level ComputationJusang Oh
Jul 18Representation LearningOn the Rankability of Visual EmbeddingsMintaek Lim
LLMsKV-efficient language models: MLA and sliding window attentionYejoon Lee