LG AI Research has unveiled VinQA, a study that generates answers by strategically combining text and visual elements.
LG AI Research has unveiled 'VinQA,' a multimodal document QA study that generates long-form answers by appropriately placing text and visual elements together.
xHC has been proposed as a technique that expands the Residual Stream into multiple streams to increase a model's memory capacity and intelligence.
Microsoft has released PIKE-RAG, which overcomes the limitations of RAG by strengthening specialized knowledge extraction and reasoning capabilities.
A study has found that quantizing Qwen2.5-Coder to 4-bit results in almost no degradation in code generation performance.
Starbucks is developing its own AI tools by leveraging AI-assisted development to cut costs on existing software.
x.AI has launched Grok Build, an AI coding agent that supports code editing and shell command execution in a terminal (TUI) environment.
Inkling, a 975B-scale open-weight multimodal model designed for fine-tuning, has been released.
A public leaderboard has launched to compete on LoRA fine-tuning speed under identical hardware and dataset conditions.
During Elon Musk's lawsuit against OpenAI, an email surfaced revealing that Sam Altman had planned to release a GPT-3-class model capable of running locally.
OpenAI backported a change reducing the model context size from 372k to 272k for the Codex 0.144 release.
An open-source template based on FastAPI and Next.js for building AI agent and RAG applications has been released.
LLM-based synthetic consumers are emerging as an alternative to market research by reproducing real customers' preferences and decision-making patterns with high accuracy.
The performance gap for open models is narrowing and inference costs are plummeting, shifting the center of competition to the agentic layer.
An open model with 386M parameters has been released that implements memory through a trained vector bank instead of a long context window.
OpenWiki, a CLI tool that integrates codebases and local knowledge to manage wikis for AI agents, has been released.
Cohere Labs has released co/plot, a tool that enables fast visualization of research data while maintaining accuracy.
Starting July 20, Claude Fable 5 will be included in Max and Team Premium plans, while Pro and Team Standard users can access it via credits.
LenVM optimizes LLM inference efficiency and performance by predicting the remaining generation length at the token level.
Vercel Workflows has introduced a feature that lets you pin run state, queue, and output streams to a specific region.
A feature has been added that lets Vercel team owners clear the Remote Cache in one go to remove poisoned artifacts.
FuriosaAI's RNGD accelerator will be offered as NPUaaS on the Samsung Cloud Platform, increasing enterprises' access to AI infrastructure.
A human-annotated LVSum benchmark has been released to evaluate temporal accuracy in long video summarization.
xAI has launched a Grok add-in that supports natural language data analysis and formula writing within Excel.
AI Briefing