Autoregressive vision-language models achieve strong multimodal understanding, but their sequential decoding becomes increasingly inefficient as multimodal responses grow longer. Adapting AR-VLMs to diffusion-style block decoding is promising, yet fixed refinement blocks struggle with sharply varying multimodal uncertainty in OCR, chart, document, and reasoning-heavy outputs. Binary mask-or-token states further limit local refinement, as masking discards partial evidence while premature commitment can corrupt the causal prefix. We present FLUID-V, a lightweight AR-to-diffusion adaptation framework for reliable local multimodal refinement. FLUID-V selects refinement spans with elastic horizons and represents unresolved positions as evolving soft semantic mixtures in embedding space. Only confident predictions are committed to the persistent KV cache, preserving causal multimodal conditioning and autoregressive compatibility. Experiments with OpenPangu-VL-7B and 100K self-distilled samples show competitive performance across diverse multimodal benchmarks, with ablations confirming the importance of adaptive horizons and stateful intermediate representations. Our code is available at https://github.com/Oli-lab-nun/FLUID-V.

Diffusion models promise efficient parallel text generation but rely on bidirectional attention, creating a structural mismatch with pre-trained Autoregressive (AR) models. This incompatibility precludes reusing robust AR priors, necessitating prohibitive pre-training from scratch. To bridge this gap, we propose FLUID, a framework that efficiently adapts AR backbones to the diffusion paradigm. By enforcing Strictly Causal Alignment, FLUID enables seamless initialization from standard GPT-style checkpoints, circumventing the need for massive pre-training. Furthermore, we introduce Elastic Horizons, an entropy-driven mechanism that dynamically modulates denoising strides based on local information density rather than fixed schedules. Experiments demonstrate that FLUID achieves state-of-the-art performance while reducing training costs by orders of magnitude, effectively reconciling established AR foundations with efficient parallel generation. Our code is available at https://huggingface.co/MYTH-Lab/FLUID.

Mixing languages are widely used in social media, especially in multilingual societies like India. Detecting the emotions contained in these languages, which is of great significance to the development of society and political trends. In this paper, we propose an ensemble of pesudo-label based Bert model and TFIDF based SGDClassifier model to identify the sentiments of Hindi-English (Hi-En) code-mixed data. The ensemble model combines the strengths of rich semantic information from the Bert model and word frequency information from the probabilistic ngram model to predict the sentiment of a given code-mixed tweet. Finally our team got an average F1 score of 0.731 on the final leaderboard,and our codalab username is will_go.