LLM

Are language agents capable of planning in more complex yet realistic settings, closer to those in which humans operate?

The success of AI models relies on the availability of large, diverse, and high-quality datasets, which can be challenging to obtain due to data scarcity, privacy concerns, and high costs.

We find task balancing and enrichment techniques are overlooked but critical to effective instruction tuning, and in particular, training with mixed prompt settings (zero-shot, few-shot, and chain-of-thought) actually yields stronger (2%+) performance in all settings.

PaLM-E-562B, with 562B parameters, is the largest vision-language model reported and achieves state-of-the-art performance on OK-VQA without task-specific finetuning.

Retro enhances auto-regressive language models by conditioning on document chunks retrieved from a large corpus.

能否利用自回归模型生成的分布重建input?

通过与多个角色进行多轮自我协作，将单个大型语言模型转化为认知协同者。