Self-Improving VLM Judges Without Human Annotation
https://www.docswell.com/s/DeepLearning2023/KN9YRQ-2025-12-19-161307
カテゴリー: dls-2025
Self-Improving VLM Judges Without Human Annotation
【DL輪読会】Self-Improving VLM Judges Without Human Annotation by @DeepLearning2023
“ROBOTOUILLE: AN ASYNCHRONOUS PLANNING BENCHMARK FOR LLM AGENTS”
【DL輪読会】"ROBOTOUILLE: AN ASYNCHRONOUS PLANNING BENCHMARK FOR LLM AGENTS" by more
Next-Latent Prediction Transformers Learn Compact World Models
【DL輪読会】Next-Latent Prediction Transformers Learn Compact World Models by @DeepLearnin more
Training-Free Activation Sparsity in Large Language Models
【DL輪読会】Training-Free Activation Sparsity in Large Language Models by @DeepLearning202 more
Mixture of Contexts for Long Video Generation
【DL輪読会】Mixture of Contexts for Long Video Generation by @DeepLearning2023
World-in-World: World Models in a Closed-Loop World
【DL輪読会】World-in-World: World Models in a Closed-Loop World by @DeepLearning2023
Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think
【DL輪読会】Representation Alignment for Generation: Training Diffusion Transformers Is Ea more
Why Diffusion Models Don’t Memorize: The Role of Implicit Dynamical Regularization in Training
【DL輪読会】 Why Diffusion Models Don’t Memorize: The Role of Implicit Dynamical Regulariz more
Gradient Estimation with Discrete Stein Operators
【DL輪読会】Gradient Estimation with Discrete Stein Operators by @DeepLearning2023