Recent advances in Multi-modal Large Language Models (MLLMs), such as LLaVA-series models, are driven by massive machine-generated instruction-following data tuning. Such automatic instruction collection pipelines, however, inadvertently introduce significant variability in data quality. This paper introduces a novel instruction curation algorithm, derived from two unique perspectives, human and LLM preference alignment, to compress this vast corpus of machine-generated multimodal instructions to a compact and high-quality form: (i) For human preference alignment, we have collected a machine-generated multimodal instruction dataset and established a comprehensive set of both subjective and objective criteria to guide the data quality assessment critically from human experts. By doing so, a reward model was trained on the annotated dataset to internalize the nuanced human understanding of instruction alignment. (ii) For LLM preference alignment, given the instruction selected by the reward model, we propose leveraging the inner LLM used in MLLM to align the writing style of visual instructions with that of the inner LLM itself, resulting in LLM-aligned instruction improvement. Extensive experiments demonstrate that we can maintain or even improve model performance by compressing synthetic multimodal instructions by up to 90%. Impressively, by aggressively reducing the training instructions from 158k to 14k (9× smaller), our model consistently outperforms its full-size dataset counterpart across various MLLM benchmarks. Our project is available at https://github.com/DCDmllm/Align2LLaVA.

Multimodal aspect-based sentiment analysis (MABSA) aims to extract the aspect terms from text and image pairs, and then analyze their corresponding sentiment. Recent studies typically use either a pipeline method or a unified transformer based on a cross-attention mechanism. However, these methods fail to explicitly and effectively incorporate the alignment between text and image. Supervised finetuning of the universal transformers for MABSA still requires a certain number of aligned image-text pairs. This study proposes a dual-encoder transformer with cross-modal alignment (DTCA). Two auxiliary tasks, including text-only extraction and text-patch alignment are introduced to enhance cross-attention performance. To align text and image, we propose an unsupervised approach which minimizes the Wasserstein distance between both modalities, forcing both encoders to produce more appropriate representations for the final extraction. Experimental results on two benchmarks demonstrate that DTCA consistently outperforms existing methods.

Data sharing restrictions are common in NLP datasets. The purpose of this task is to develop a model trained in a source domain to make predictions for a target domain with related domain data. To address the issue, the organizers provided the models that fine-tuned a large number of source domain data on pre-trained models and the dev data for participants. But the source domain data was not distributed. This paper describes the provided model to the NER (Name entity recognition) task and the ways to develop the model. As a little data provided, pre-trained models are suitable to solve the cross-domain tasks. The models fine-tuned by large number of another domain could be effective in new domain because the task had no change.