Blog
데이터 과학, 인공지능, 딥러닝에 관한 이야기
186개 중 1-12번째 포스트
![[논문 리뷰] EvoSkill: Automated Skill Discovery for Multi-Agent Systems](/assets/images/blog/20260307-paper-2603-02766-evoskill-automated-skill-disco.jpg)
[논문 리뷰] EvoSkill: Automated Skill Discovery for Multi-Agent Systems
Coding agents are increasingly used as general-purpose problem solvers, but their flexibility does not by itself confer the domain expertise needed for specialized tasks. Recent work addresses this th...
![[논문 리뷰] Evaluating Theory of Mind and Internal Beliefs in LLM-Based Multi-Agent Systems](/assets/images/blog/20260307-paper-2603-00142-evaluating-theory-of-mind-and-.jpg)
[논문 리뷰] Evaluating Theory of Mind and Internal Beliefs in LLM-Based Multi-Agent Systems
LLM-based MAS are gaining popularity due to their potential for collaborative problem-solving enhanced by advances in natural language comprehension, reasoning, and planning. Research in Theory of Min...
![[논문 리뷰] ParamMem: Augmenting Language Agents with Parametric Reflective Memory](/assets/images/blog/20260307-paper-2602-23320-parammem-augmenting-language-a.jpg)
[논문 리뷰] ParamMem: Augmenting Language Agents with Parametric Reflective Memory
Self-reflection enables language agents to iteratively refine solutions, yet often produces repetitive outputs that limit reasoning performance. Recent studies have attempted to address this limitatio...
![[논문 리뷰] Large-scale online deanonymization with LLMs](/assets/images/blog/20260307-paper-2602-16800-large-scale-online-deanonymiza.jpg)
[논문 리뷰] Large-scale online deanonymization with LLMs
We show that large language models can be used to perform at-scale deanonymization. With full Internet access, our agent can re-identify Hacker News users and Anthropic Interviewer participants at hig...
![[논문 리뷰] From SGD to Spectra: A Theory of Neural Network Weight Dynamics](/assets/images/blog/20260307-paper-2507-12709-from-sgd-to-spectra-a-theory-o.jpg)
[논문 리뷰] From SGD to Spectra: A Theory of Neural Network Weight Dynamics
Deep neural networks have revolutionized machine learning, yet their training dynamics remain theoretically unclear-we develop a continuous-time, matrix-valued stochastic differential equation (SDE) f...
![[논문 리뷰] ParamMem: Augmenting Language Agents with Parametric Reflective Memory](/assets/images/blog/20260305-paper-2602-23320-parammem-augmenting-language-a.jpg)
[논문 리뷰] ParamMem: Augmenting Language Agents with Parametric Reflective Memory
Self-reflection enables language agents to iteratively refine solutions, yet often produces repetitive outputs that limit reasoning performance. Recent studies have attempted to address this limitatio...
![[논문 리뷰] Stress Testing Deliberative Alignment for Anti-Scheming Training](/assets/images/blog/20260305-paper-2509-15541-stress-testing-deliberative-al.jpg)
[논문 리뷰] Stress Testing Deliberative Alignment for Anti-Scheming Training
Highly capable AI systems could secretly pursue misaligned goals -- what we call "scheming". Because a scheming AI would deliberately try to hide its misaligned goals and actions, measuring and mitiga...
![[논문 리뷰] Vectorizing the Trie: Efficient Constrained Decoding for LLM-based Generative Retrieval on Accelerators](/assets/images/blog/20260301-paper-2602-22647-vectorizing-the-trie-efficient.jpg)
[논문 리뷰] Vectorizing the Trie: Efficient Constrained Decoding for LLM-based Generative Retrieval on Accelerators
Generative retrieval has emerged as a powerful paradigm for LLM-based recommendation. However, industrial recommender systems often benefit from restricting the output space to a constrained subset of...
![[논문 리뷰] A Very Big Video Reasoning Suite](/assets/images/blog/20260301-paper-2602-20159-a-very-big-video-reasoning-sui.jpg)
[논문 리뷰] A Very Big Video Reasoning Suite
Rapid progress in video models has largely focused on visual quality, leaving their reasoning capabilities underexplored. Video reasoning grounds intelligence in spatiotemporally consistent visual env...
![[논문 리뷰] From Blind Spots to Gains: Diagnostic-Driven Iterative Training for Large Multimodal Models](/assets/images/blog/20260228-paper-2602-22859-from-blind-spots-to-gains-diag.jpg)
[논문 리뷰] From Blind Spots to Gains: Diagnostic-Driven Iterative Training for Large Multimodal Models
As Large Multimodal Models (LMMs) scale up and reinforcement learning (RL) methods mature, LMMs have made notable progress in complex reasoning and decision making. Yet training still relies on static...
![[논문 리뷰] MiroFlow: Towards High-Performance and Robust Open-Source Agent Framework for General Deep Research Tasks](/assets/images/blog/20260228-paper-2602-22808-miroflow-towards-high-performa.jpg)
[논문 리뷰] MiroFlow: Towards High-Performance and Robust Open-Source Agent Framework for General Deep Research Tasks
Despite the remarkable progress of large language models (LLMs), the capabilities of standalone LLMs have begun to plateau when tackling real-world, complex tasks that require interaction with externa...
![[논문 리뷰] Search More, Think Less: Rethinking Long-Horizon Agentic Search for Efficiency and Generalization](/assets/images/blog/20260228-paper-2602-22675-search-more-think-less-rethink.jpg)
[논문 리뷰] Search More, Think Less: Rethinking Long-Horizon Agentic Search for Efficiency and Generalization
Recent deep research agents primarily improve performance by scaling reasoning depth, but this leads to high inference cost and latency in search-intensive scenarios. Moreover, generalization across h...