Recent Multimodal Large Language Models (MLLMs) support retrieval-augmented generation (RAG) for document question answering (QA), yet it remains unclear how effectively they use the provided evidence during answer generation. In this work, we conduct a controlled empirical study of 7 popular MLLMs on long multimodal m...
Debolena Basak, Digbalay Bose, Koustava Goswami et al.· Proceedings of the 32nd ACM...· 0 citations
SimLoss is proposed, a reference-free embedding-space objective for single-pass fine-grained image captioning, and results show that embedding-space supervision can recover the quality of multi-stage verification at the latency of a single-pass captioner.
Suryaansh Jain, Rahasya Barkur, G. Vishal et al.· 0 citations
SignBALANCE is proposed, whose magnitude is composition-free: it keeps the verifier sign, uses a global scale, and restores zero-mean balance via a stop-gradient per-class rescaling.
Jiamian Wang, Samyadeep Basu, Koustava Goswami et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
