Results for “charavault”

Runs the open-source LLaVA vision-language model for image understanding, captioning, visual question answering, and multi-turn image conversations, including setup, inference, and training guidance.

Benchmarks large audio-language models against adversarial audio attacks using the CAA dataset, computing WER, ROUGE-L, cosine similarity, and coherence scores to assess robustness in conversational settings.