주제로 연구실 찾기

1. KAIST·컴퓨터 비전

   노용만IVY Lab

   관련도관련 논문 10편비중 13%연구 분야에 명시
   • 2026A Causal Lens on Non-RGB Vision Sensor Understanding in Vision Language ModelsIEEE TIP
   • 2026GenRecal: Generation after Recalibration from Large to Small Vision-Language ModelsECCV
   • 2026MAD: Modality-Adaptive Decoding for Mitigating Cross-Modal Hallucinations in Multimodal Large Language ModelsCVPR
2. KAIST·컴퓨터 비전

   심현정CVML Lab

   관련도관련 논문 8편비중 15%
   • 2026Blind to Position, Biased in Language: Probing Mid-Layer Representational Bias in Vision-Language Encoders for Zero-Shot Language-Grounded Spatial UnderstandingECCV
   • 2026Interpretable Debiasing of Vision-Language Models for Social FairnessCVPR
   • 2026What 'Not' to Detect: Negation-Aware VLMs via Structured Reasoning and Token MergingICLR
3. KAIST·컴퓨터 비전

   주재걸DAVIAN Lab

   관련도관련 논문 9편비중 7%
   • 20263D HAMSTER: Bridging Planning and Control in Hierarchical Vision Language Action Models through 3D Trajectory GuidanceIROS
   • 2026ACG: Action Coherence Guidance for Flow-Based Vision-Language-Action ModelsICRA
   • 2026MM-SeR: Multimodal Self-Refinement for Lightweight Image CaptioningCVPR
4. KAIST·컴퓨터 비전

   오태현AMI Lab

   관련도관련 논문 8편비중 8%
   • 2026Aligning What Vision-Language Models See and Perceive with Adaptive Information FlowCVPR
   • 2026CLAY: Conditional Visual Similarity Modulation in Vision-Language Embedding SpaceCVPR
   • 2026CLIP-Actor-X: Text-driven 4D Human Avatar Generation via Cross-modal Synthesis-through-OptimizationTPAMI
5. KAIST·컴퓨터 비전

   김승룡CV Lab

   관련도관련 논문 7편비중 6%
   • 20264DP-QA: Scalable QA for 4D Perception in Vision Language ModelsCVPR
   • 2026Entropy-Gradient Grounding: Training-Free Evidence Retrieval in Vision-Language ModelsECCV
   • 2026Imagine3D-LLM: Teaching MLLMs to Imagine 3D Scenes Before AnsweringNeurIPS
6. KAIST·컴퓨터 비전

   김현우MLV Lab

   관련도관련 논문 6편비중 8%연구 분야에 명시
   • 2026Context Blindness in DPO: Mitigating Object Hallucination in MLLMs via Context-Calibrated Preference OptimizationECCV
   • 2026MoE-GRPO: Optimizing Mixture-of-Experts via Reinforcement Learning in Vision-Language ModelsCVPR
   • 2026Transferable Model-agnostic Vision-Language Model Adaptation for Efficient Weak-to-Strong GeneralizationAAAI
7. KAIST·머신러닝

   신진우ALIN Lab

   관련도관련 논문 8편비중 6%
   • 2026Contrastive Representation Regularization for Vision-Language-Action ModelsICML
   • 2026Dual-Stream Diffusion for World-Model Augmented Vision-Language-Action ModelICML
   • 2026HAMLET: Switch Your Vision-Language-Action Model into a History-Aware PolicyICLR
8. KAIST·자연어처리·LLM

   임경태MLP Lab

   관련도관련 논문 5편비중 13%연구 분야에 명시
   • 2026Beyond Black Boxes: Explainable Face Anti-Spoofing with Vision-Language ModelsScientific Reports
   • 2026Planning Before Reasoning: A Vision-Language Dataset for Structured Stepwise Reasoning in Document UnderstandingEMNLP Findings
   • 2025VLR-Bench: Multilingual Benchmark Dataset for Vision-Language Retrieval Augmented GenerationCOLING
9. KAIST·머신러닝

   황성주MLAI Lab

   관련도관련 논문 5편비중 5%
   • 2026ES-Merging: Biological MLLM Merging via Embedding Space SignalsNeurIPS
   • 2026MultihopSpatial: Multi-hop Compositional Spatial Reasoning Benchmark for Vision-Language ModelECCV
   • 2026Multimodal Prompt Optimization: Why Not Leverage Multiple Modalities for MLLMsICLR
10. KAIST·머신러닝

    양은호MLI Lab

    관련도관련 논문 5편비중 6%
    • 2026CollabVR: Collaborative Video Reasoning with Vision-Language and Video Generation ModelsNeurIPS
    • 2026MM-JudgeBias: A Benchmark for Evaluating Compositional Biases in MLLM-as-a-JudgeACL
    • 2025Unveiling the Response of Large Vision-Language Models to Visually Absent TokensEMNLP
11. KAIST·머신러닝

    윤세영OSI Lab

    관련도관련 논문 5편비중 3%
    • 2026Stage-wise Attention-Guided Region Sequencing for Adversarial Attacks on Large Vision-Language ModelsNeurIPS
    • 2026Route Guided CBF-QP Repair for Safe Vision Language Action ModelsRSS
    • 2025Bayesian Principles Improve Prompt Learning in Vision-Language ModelsAISTATS
12. KAIST·컴퓨터 비전

    김준모SIIT Lab

    관련도관련 논문 4편비중 5%
    • 2026MLLMCLIP: Feature-Level Distillation of MLLM for Robust Vision-Language RepresentationsEMNLP
    • 2026Beyond Semantics: Disentangling Information Scope in Sparse Autoencoders for CLIPCVPR Findings
    • 2026IWP: Token Pruning as Implicit Weight Pruning in Large Vision Language ModelsEMNLP Findings
13. KAIST·데이터 마이닝

    이재길Data Mining Lab

    관련도관련 논문 4편비중 10%
    • 2026CURE: Cumulative Knowledge Reuse for Efficient Device-Server Hybrid Inference in Vision-Language ModelsECCV
    • 2026See and Fix the Flaws: Enabling VLMs and Diffusion Models to Comprehend Visual Artifacts via Agentic Data SynthesisCVPR
    • 2025RA-TTA: Retrieval-Augmented Test-Time Adaptation for Vision-Language ModelsICLR
14. KAIST·머신러닝

    문재균

    관련도관련 논문 3편비중 15%
    • 2026Identifying Robust Neural Pathways: Few-Shot Adversarial Mask Tuning for Vision-Language ModelsICLR
    • 2026ProLoG: Hybrid Prompt and LoRA Based Adaptation of Vision-Language Models for OOD GeneralizationAAAI
    • 2026Policy-Only Power Sampling for Vision-Language-Action ControlICML
15. KAIST·머신러닝

    오성준STAI Lab

    관련도관련 논문 3편비중 7%
    • 2026CLIP Behaves like a Bag-of-Words Model Cross-modally but not Uni-modallyICLR
    • 2026CRISP: Compositional Reasoning over Images via Stackable Programs for VLMsNeurIPS
    • 2026CLIP Models Generalize Less Than Compositional Benchmarks SuggestICML
16. KAIST·컴퓨터 비전

    성민혁Visual AI Group

    관련도관련 논문 3편비중 7%
    • 2026Retrieval Heads Meet Vision: Uncovering How VLMs Locate and Extract Visual InformationNeurIPS
    • 2026Token Warping Helps MLLMs Look from Nearby ViewpointsCVPR
    • 2025Perspective-Aware Reasoning in Vision-Language Models via Mental Imagery SimulationICCV
17. KAIST·AI for Science

    예종철BISPL

    관련도관련 논문 4편비중 3%
    • 2026Read like a radiologist: Efficient vision-language model for 3d medical imaging interpretationMedical Image Analysis
    • 2025Free2Guide: Training-Free Text-to-Video Alignment using Image LVLMICCV
    • 2024Improving Medical Speech-to-Text Accuracy using Vision-Language Pre-training ModelsIEEE JBHI
18. KAIST·데이터 마이닝

    박찬영DSAIL

    관련도관련 논문 3편비중 3%
    • 2026Why and When Visual Token Pruning Fails? A Study on Relevant Visual Information Shift in MLLMs DecodingECCV
    • 2026Test-Time Training for Visual Foresight Vision-Language-Action ModelsICML
    • 2024Vision Language Model is NOT All You Need: Augmentation Strategies for Molecule Language ModelsCIKM
19. KAIST·자연어처리·LLM

    오혜연U&I Lab

    관련도관련 논문 3편비중 4%
    • 2026World in a Frame: Understanding Culture Mixing as a New Challenge for Vision-Language ModelsCVPR
    • 2025WorldCuisines: A Massive-Scale Benchmark for Multilingual and Multicultural Visual Question Answering on Global CuisinesNAACL
    • 2025When Tom Eats Kimchi: Evaluating Cultural Bias of Multimodal Large Language Models in Cultural Mixture ContextsNAACL
20. KAIST·컴퓨터 비전

    홍승훈Vision and Learning Lab

    관련도관련 논문 2편비중 5%
    • 2026Multi-View Relational Distillation for Spatial Reasoning with Vision-Language ModelsNeurIPS
    • 2026When Sinks Help or Hurt: Unified Framework for Attention Sink in Large Vision-Language ModelsECCV
21. KAIST·컴퓨터 비전

    윤국진VI Lab

    관련도관련 논문 2편비중 2%
    • 2026Multimodal Distribution Matching for Vision-Language Dataset DistillationCVPR
    • 2026Rank-Aware Hyperbolic Alignment for Vision-Language Dataset DistillationECCV
22. KAIST·자연어처리·LLM

    서민준LK Lab

    관련도관련 논문 3편비중 5%
    • 2025How Does Vision-Language Adaptation Impact the Safety of Vision Language Models?ICLR
    • 2025Dynamic VLM-Guided Negative Prompting for Diffusion ModelsNeurIPS
    • 2024Prometheus-Vision: Vision-Language Model as a Judge for Fine-Grained EvaluationACL Findings
23. KAIST·머신러닝

    김기응AIPR Lab

    관련도관련 논문 2편비중 7%
    • 2026PRISM-VLM: A Multi-Axis Discriminative Benchmark for Compact Vision-Language ModelsEMNLP Findings
    • 2025Dynamic VLM-Guided Negative Prompting for Diffusion ModelsNeurIPS
24. KAIST·음성·오디오

    정준선Multimodal AI Lab

    관련도관련 논문 2편비중 2%
    • 2026Hearing and Seeing Through CLIP: A Framework for Self-Supervised Sound Source LocalizationIJCV
    • 2024Can CLIP Help Sound Source Localization?WACV
25. KAIST·컴퓨터 비전

    김승욱GLOW Lab

    관련도관련 논문 1편비중 7%
    • 2026Compositional SVG Generation via VLM-Driven Hierarchical Semantic ParsingEMNLP
26. KAIST·로보틱스

    박대형RIRO Lab

    관련도관련 논문 1편비중 7%
    • 2026C2F-SPACE: Coarse-to-Fine Space Grounding for Spatial Instructions using Vision-Language ModelsIntelligent Service Robotics
27. KAIST·AI 시스템

    유민수VIA Research Group

    관련도관련 논문 1편비중 4%
    • 2026Exploring KV Cache Quantization in Multimodal Large Language Model InferenceIEEE Computer Architecture Letters
28. KAIST·머신러닝

    정혜원IIDS Lab

    관련도관련 논문 1편비중 4%
    • 2026InfCLIP: Unified Data Valuation for CLIP Pretraining via Influence-Inspired ScoringNeurIPS
29. KAIST·AI 시스템

    김주영CAST Lab

    관련도관련 논문 1편비중 1%
    • 2026Tri-Oracle: A 17.78 μJ/Token Vision-Language Model Accelerator with Token-Attention-Weight Redundancy PredictionISSCC
30. KAIST·인간-AI 상호작용

    김주호KIXLAB

    관련도관련 논문 1편비중 1%
    • 2026CANVAS: A Benchmark for Vision-Language Models on Tool-Based User Interface DesignAAAI
31. KAIST·데이터 마이닝

    신기정Data Mining Lab

    관련도관련 논문 1편비중 1%
    • 2026Pre-Deployment Advertisement Ranking under Data Scarcity via Context-Aware Criteria Generation with VLMsACL
32. KAIST·머신러닝

    이주호SIML Lab

    관련도관련 논문 1편비중 1%
    • 2026Bridging the missing-modality gap: improving text-only calibration of vision language modelsICLR
33. KAIST·머신러닝

    문일철AAI Lab

    관련도관련 논문 1편비중 3%
    • 2024Make Prompts Adaptable : Bayesian Modeling for Vision-Language Prompt Learning with Data-Dependent PriorAAAI
34. KAIST·머신러닝

    이기민

    관련도관련 논문 1편비중 2%
    • 2024By My Eyes: Grounding Multimodal Large Language Models with Sensor Data via Visual PromptingEMNLP
35. KAIST·AI for Science

    최윤재

    관련도관련 논문 1편비중 2%
    • 2024Vision-Language Generative Model for View-Specific Chest X-ray GenerationCHIL

2026 논문이 아직 반영되지 않은 연구실은 결과에서 불리할 수 있어요.