Compositionality is still a challenging problem.

Large language models are now tuned to align with the goals of their creators, namely to be "helpful and harmless." These models should respond helpfully to user questions, but refuse to answer requests that could cause harm. However, adversarial users can construct inputs which circumvent attempts at alignment. In this work, we study adversarial alignment, and ask to what extent these models remain aligned when interacting with an adversarial user who constructs worst-case inputs (adversarial examples). These inputs are designed to cause the model to emit harmful content that would otherwise be prohibited. We show that existing NLP-based optimization attacks are insufficiently powerful to reliably attack aligned text models: even when current NLP-based attacks fail, we can find adversarial inputs with brute force.

Multimodal interleaved datasets featuring free-form interleaved sequences of images and text are crucial for training frontier large multimodal models (LMMs).

However, the integration of heterogeneous multimodal data poses a significant challenge, as confounding effects and dependencies among such heterogeneous data sources introduce unwanted variability and bias, leading to suboptimal performance of multimodal models. Therefore, it becomes crucial to normalize the low-or high-level features extracted from data modalities before their fusion takes place.