Aleena, an open-source AI agent developed by a university-based research software engineering center, aims to bridge communication gaps in research software projects by turning chaotic interactions in…
Researchers introduced LEEVLA, a new architecture for vision-language-action models that uses drift-guided dynamic prioritization to focus on task-relevant evidence, outperforming existing methods on …
Autonomous web agents face old security threats like prompt injection, and a new defense called Prismata aims to protect them using contextual least privilege. Prismata dynamically labels web content …
Researchers introduced the PS4 framework, a proxy-supervised training method for target speaker extraction in conversational AI, using a bilingual dataset of 71,771 samples. The framework achieved sec…
South Korean chip maker SK hynix set pricing for its mega US listing on Friday, aiming to raise $26.5bn in one of the world's biggest stock sales, capitalizing on the AI boom. The company, a supplier …
Chancellor Rachel Reeves will launch a City 'skills compact' on Tuesday, committing financial firms including Barclays and Lloyds to retrain thousands of workers for the AI revolution. The government-…
US artificial intelligence models were sold to Chinese entities on a US blacklist, violating export controls. The sales, which involved companies including a subsidiary of ByteDance, raise national se…
An AI 'superforecaster' designed to predict market movements, including Federal Reserve decisions, has not yet outperformed human forecasters, according to a Financial Times report. The AI's performan…
Artificial intelligence is transforming sports viewership by enabling personalized feeds, reducing reliance on expensive subscriptions and fixed schedules, according to a Financial Times report.…
OpenAI and Google have been selling AI models to Singapore-based subsidiaries of Chinese tech giants Alibaba, Baidu, and Tencent, which are blacklisted by the US government. The sales potentially viol…
Regulators are exploring the use of artificial intelligence to reduce bureaucratic red tape, aiming to streamline financial oversight and cut administrative burdens. The technology could help automate…
Headlong's production of Karel Čapek's 1920 play RUR: Rossum's Universal Robots, adapted by Ella Road, opens at Oxford's Schwarzman Centre, using research from Oxford University academics to explore t…
Researchers developed a modular speaker diarization front end combined with an adapted Qwen3-ASR-1.7B recognizer for multilingual two-speaker conversational speech, achieving a tcpMER of 17.97 on the …
Researchers introduced SQuaD-SQL, a method that uses knowledge distillation from large language models to train small language models for text-to-SQL tasks, achieving 86.9% execution accuracy on WikiS…
Researchers found that best-of-N text-to-speech evaluation is confounded by alignment between ASR verifier and evaluator families, with same-family pairs recovering 2-3× more oracle headroom than cros…
Researchers have developed a method to compress instruction prompts into a single activation vector for large language models, achieving under 2% accuracy loss compared to full prompt processing. The …
Researchers at an undisclosed institution applied Group Relative Policy Optimization (GRPO) to adapt an LLM-based automatic speech recognition (ASR) system to synthetic speech, achieving a 40% relativ…
Researchers introduced LEXIC, a lightweight eye-tracking extension that injects word-level difficulty signals into gaze-only models, achieving AUROC gains of up to +2.9 percentage points on reading co…
The ICDAR 2026 HIPE-OCRepair competition evaluated LLM-assisted OCR post-correction for historical documents in English, French, and German. Four teams submitted systems using strategies from zero-sho…
Researchers developed a two-stage system using large language models to extract fine-grained event tags from SEC 8-K filings, mapping disclosures to a 119-event taxonomy with grounded quotes. Applied …