AI Intel

Qwen/Qwen-Image-Bench has been released on Hugging Face, featuring Q-Judger, a vision-language model fine-tuned for automated evaluation of text-to-i…

A study reveals that five frontier large language models disagree on 67% of 1,000 real-world fact-check claims.

Researchers have introduced Soro, a family of lightweight Tajik-specialized conversational large language models (LLMs) and chatbots designed for rea…

Research explores fine-tuning vision-language models to understand current damage and score priority using a Quality Guard Agent, applicable to scena…

A paper introduces $E^3$-Agent, an executable and evolving agent designed for resource management in edge generative inference, addressing challenges…

A study proposes a tailorable LLM-based architecture for identifying and understanding human values embedded in text, aiming to integrate ethical con…

A paper presents DynaSchedBench, a set of calibrated dynamic scheduling benchmarks, and discusses the "observability paradox" in LLM-based scheduling…

Researchers propose ICG, a method that significantly improves cover image generation quality by combining MLLM-based prompting with personalized pref…

A paper introduces LCO, an LLM-based constraint optimization method designed to enhance the safety of agentic LLMs in real-world tasks, addressing is…

Research focuses on achieving fine-grained and within-utterance speaking style control in prompt-based text-to-speech (TTS) models.