Blog
210개 중 1-12번째 포스트
![[논문 리뷰] Intuitive physics understanding emerges from self-supervised pretraining on natural videos](/assets/images/blog/20260427-paper-2502-11831-intuitive-physics-understandin.jpg)
[논문 리뷰] Intuitive physics understanding emerges from self-supervised pretraining on natural videos
We investigate the emergence of intuitive physics understanding in general-purpose deep neural network models trained to predict masked regions in natural videos. Leveraging the violation-of-expectati...
![[논문 리뷰] Intuitive physics understanding emerges from self-supervised pretraining on natural videos](/assets/images/blog/20260405-paper-2502-11831-intuitive-physics-understandin.jpg)
[논문 리뷰] Intuitive physics understanding emerges from self-supervised pretraining on natural videos
We investigate the emergence of intuitive physics understanding in general-purpose deep neural network models trained to predict masked regions in natural videos. Leveraging the violation-of-expectati...
![[논문 리뷰] Internalizing Agency from Reflective Experience](/assets/images/blog/20260319-paper-2603-16843-internalizing-agency-from-refl.jpg)
[논문 리뷰] Internalizing Agency from Reflective Experience
Large language models are increasingly deployed as autonomous agents that must plan, act, and recover from mistakes through long-horizon interaction with environments that provide rich feedback. Howev...
![[논문 리뷰] Language Models are Injective and Hence Invertible](/assets/images/blog/20260319-paper-2510-15511-language-models-are-injective-.jpg)
[논문 리뷰] Language Models are Injective and Hence Invertible
Transformer components such as non-linear activations and normalization are inherently non-injective, suggesting that different inputs could map to the same output and prevent exact recovery of the in...
![[논문 리뷰] Attention Residuals](/assets/images/blog/20260318-paper-2603-15031-attention-residuals.jpg)
[논문 리뷰] Attention Residuals
Residual connections with PreNorm are standard in modern LLMs, yet they accumulate all layer outputs with fixed unit weights. This uniform aggregation causes uncontrolled hidden-state growth with dept...
![[논문 리뷰] Sparking Scientific Creativity via LLM-Driven Interdisciplinary Inspiration](/assets/images/blog/20260317-paper-2603-12226-sparking-scientific-creativity.jpg)
[논문 리뷰] Sparking Scientific Creativity via LLM-Driven Interdisciplinary Inspiration
Despite interdisciplinary research leading to larger and longer-term impact, most work remains confined to single-domain academic silos. Recent AI-based approaches to scientific discovery show promise...
![[논문 리뷰] Nurture-First Agent Development: Building Domain-Expert AI Agents Through Conversational Knowledge Crystallization](/assets/images/blog/20260316-paper-2603-10808-nurture-first-agent-developmen.jpg)
[논문 리뷰] Nurture-First Agent Development: Building Domain-Expert AI Agents Through Conversational Knowledge Crystallization
The emergence of large language model (LLM)-based agent frameworks has shifted the primary challenge in building domain-expert AI agents from raw capability to effective encoding of domain expertise. ...
![[논문 리뷰] Reinforced Generation of Combinatorial Structures: Ramsey Numbers](/assets/images/blog/20260316-paper-2603-09172-reinforced-generation-of-combi.jpg)
[논문 리뷰] Reinforced Generation of Combinatorial Structures: Ramsey Numbers
We present improved lower bounds for five classical Ramsey numbers: $\mathbf{R}(3, 13)$ is increased from $60$ to $61$, $\mathbf{R}(3, 18)$ from $99$ to $100$, $\mathbf{R}(4, 13)$ from $138$ to $139$,...
![[논문 리뷰] LLM2Vec-Gen: Generative Embeddings from Large Language Models](/assets/images/blog/20260315-paper-2603-10913-llm2vec-gen-generative-embeddi.jpg)
[논문 리뷰] LLM2Vec-Gen: Generative Embeddings from Large Language Models
LLM-based text embedders typically encode the semantic content of their input. However, embedding tasks require mapping diverse inputs to similar outputs. Typically, this input-output is addressed by ...
![[논문 리뷰] OpenClaw-RL: Train Any Agent Simply by Talking](/assets/images/blog/20260315-paper-2603-10165-openclaw-rl-train-any-agent-si.jpg)
[논문 리뷰] OpenClaw-RL: Train Any Agent Simply by Talking
Every agent interaction generates a next-state signal, namely the user reply, tool output, terminal or GUI state change that follows each action, yet no existing agentic RL system recovers it as a liv...
![[논문 리뷰] LLM2Vec-Gen: Generative Embeddings from Large Language Models](/assets/images/blog/20260314-paper-2603-10913-llm2vec-gen-generative-embeddi.jpg)
[논문 리뷰] LLM2Vec-Gen: Generative Embeddings from Large Language Models
LLM-based text embedders typically encode the semantic content of their input. However, embedding tasks require mapping diverse inputs to similar outputs. Typically, this input-output is addressed by ...
![[논문 리뷰] Nurture-First Agent Development: Building Domain-Expert AI Agents Through Conversational Knowledge Crystallization](/assets/images/blog/20260312-paper-2603-10808-nurture-first-agent-developmen.jpg)
[논문 리뷰] Nurture-First Agent Development: Building Domain-Expert AI Agents Through Conversational Knowledge Crystallization
The emergence of large language model (LLM)-based agent frameworks has shifted the primary challenge in building domain-expert AI agents from raw capability to effective encoding of domain expertise. ...