New Delhi: NDTV has achieved a significant milestone in applied artificial intelligence, with its research paper accepted at ACM SIGIR 2026 – widely regarded as the world’s leading conference in search and information retrieval.
The paper, ‘All the News That Fits in Bits: Learned Rotation-Aware Binary Projections for Efficient News Retrieval at NDTV’, focuses on a challenge that sits at the core of modern news platforms – how to find and surface the right information instantly from a rapidly expanding universe of content.
As newsrooms scale and archives grow, delivering relevant stories in real time – with speed, accuracy, and efficiency – becomes increasingly complex. NDTV’s work offers a more effective way to do this, making search significantly faster and more cost-efficient without compromising quality.
The system has been tested under real-world newsroom conditions, where editorial teams found the results to be as reliable and relevant as those of existing approaches.
The acceptance is particularly notable given the selectivity of SIGIR’s Industry Track, where only a small number of papers are chosen globally from submissions by leading technology companies and research institutions. NDTV becomes the first Indian media company to be selected in this category.
Ritwick Ghosh, lead author of the paper and Engineer, Machine Learning at NDTV, said, ‘The challenge was simple to describe but difficult to solve – how do you search through a massive and constantly growing news library in milliseconds, without losing relevance? This work is about making that possible in a way that is fast, efficient, and dependable in real newsroom conditions.’
Rohan Tyagi, Chief Product Officer, NDTV Digital, added, ‘As our content and audiences scale, building intelligent systems like these becomes essential. This is about ensuring that relevance is delivered instantly, while keeping the experience seamless and efficient.’
This recognition places NDTV among a small group of global organisations contributing to advances in AI-driven search and discovery, and strengthens its position as a newsroom building technology grounded in real-world use.
