An estimated 1 billion people worldwide live with vision impairment, yet current vision-language models (VLMs) produce descriptions too vague for safe navigation by blind and low-vision (BLV) users. Large VLMs can generate high-quality audio-description-compliant narrations but cannot run on mobile devices; small VLMs offer competitive latency but lack spatial detail, directional cues, and hazard awareness for navigational assistance. We present Smol-VL-BLV, a compact VLM for blind and low-vision users that closes this gap using a 500M decoder transformer model and two post-training mechanisms: (1) teacher-student distillation and (2) Group Relative Policy Optimisation (GRPO) with a composite BLV reward targeting directional language, metric distances, and hazard detection. Because multi-stage post-training can induce catastrophic forgetting, we add a lightweight finetuning stage after the last stage GRPO finetuning to recover general descriptive quality while preserving BLV-specific spatial grounding. Our best model substantially outperforms the baseline across various benchmarks, including tasks: VQA, BLV captioning, OCR, and real-time observation. Compared with the baseline for relative improvement, it improves the Spatial score gain of 19.3%, and the Social score gain of 14.8%. It also increases OCR-Bench by 101.5%, and raises TextVQA accuracy by 44.2%. These results show that BLV-focused post-training improves both accessibility-specific spatial grounding and general visual-text reasoning. Deployed on a mid-range Android smartphone via Mixed-Precision Quantization, the model remains under 450 MB and generates descriptions in real time, demonstrating practical viability for on-device BLV assistance. Our model, dataset, and code will be publicly released.

Story building is an important part of language and overall development of a child. Developing an interactive and artificial intelligence (AI) based solution to create stories for children is an open and challenging problem. Methods combining large language models (LLMs) and knowledge graphs (KGs) have further enabled high quality and coherent story generation. In this work, we present a platform, Story Yarn, developed for interactive story creation for children. We customise a KG, using children stories, which captures relationships between components of stories. This customised KG is then used along with LLM to collaboratively create a story. We have also built a simple app to facilitate user interaction. This platform can aid the creative development of children, and can be used at home or in schools.