Large Language Model
RL-GPT: Integrating Reinforcement Learning and Code-as-policy
Additionally, it achieves good performance on designated MineDojo tasks.
OmniTokenizer: A Joint Image-Video Tokenizer for Visual Generation
Tokenizer, serving as a translator to map the intricate visual data into a compact latent space, lies at the core of visual generative models.
Uncovering the Strategic Reasoning Limitations of LLMs via Game-Theoretic Evaluations
As Large Language Models (LLMs) are integrated into critical real-world applications, their strategic and logical reasoning abilities are increasingly crucial.
Rethinking Memory and Communication Costs for Efficient Data Parallel Training of Large Language Models
Recently, various strategies for distributed training of large language models (LLMs) have been proposed.
GPT: An Advanced Multimodal, Multitask Framework for Motion Comprehension and Generation
GPT can handle core motion comprehension and generation tasks, including text-to-motion, motion-to-text, music-to-dance, dance-to-music, motion prediction, and motion in-between.
FineCLIP: Self-distilled Region-based CLIP for Better Fine-grained Understanding
However, CLIP struggles with dense prediction tasks due to the poor grasp of the fine-grained details.
Emmanuel Abbe
Can Transformers predict new syllogisms by composing established ones?
AlphaMath Almost Zero: Process Supervision without Process
LLMs possess a vast reservoir of knowledge, which remains under-utilized in current finetuning-based approaches.