AI News Archive - Browse Page 131 of 282
Browse AI news articles covering LLMs, tools, research, and industry trends
CoCoNuT paradigm expands residual stream for latent‑space, multi‑path reasoning
The promise of latent-space reasoning is irresistible: think in parallel, explore multiple paths, never prematurely commit.
OmniMem adds modality-aware memory allocation for audio‑visual LLMs
Every AI model has a memory limit, but audio-visual ones face a uniquely stupid problem.
AI agents solve neuroscience pipeline tasks on datasets larger than benchmarks
Forget the tidy benchmarks. AI is being thrown against actual scientific work now, with messy data and no clear finish line.
ML models predict World Cup outcomes, but miss draws, capture team strength
Machine learning models crave a simple fight. Give them a World Cup match, and they'll happily pick the stronger side. But a draw?
MedicalRec releases MedicalRec-Bench: 5,000+ entries for medical image classification
Medical AI research is a mess of contradictory claims, and a new dataset finally shows how deep the problem runs.
PathoSage Introduces Three‑Stage Framework for Patch‑Level Pathology Reasoning
AI systems for diagnosing tissue samples are famously unreliable. They hallucinate, get confused, and often miss the point entirely.
Apple unveils third‑gen foundation model, AFM 3 Cloud shows 36% boost
For years, Apple's AI was publicly seen as a laggard. That story ended Tuesday. The third-generation AFM 3 Cloud arrived as a wholesale correction,...
NVFP4 recipe speeds JAX/MaxText training on NVIDIA Blackwell and Rubin
Training large language models at 4-bit precision without sacrificing accuracy has long seemed like a zero-sum game, until now.
Weaker LLMs Accidentally Delete Content, Shrinking Documents Over Time
Weak AI is clumsy. Strong AI is treacherous. A feeble language model will just erase things.
Four New Specific Techniques to Boost Productivity with Claude Code
Coding agents like Claude Code hold immense potential, but most developers leave that potential on the table.
LangChain Emergency Helpline Uses AssemblyAI WebSocket for Live STT
The caller’s voice arrives in raw audio chunks, each one a fragment of a crisis. Milliseconds matter.
Jensen Huang sees token market segmenting into distinct value tiers
Forget bulk commodities. Nvidia CEO Jensen Huang sees the AI market fracturing. Every digital sliver of processing now carries a price tag directly...
OpenAI to revamp ChatGPT, shift to business customers, rival Anth
The chatbot era is over. At least, that’s the conviction now driving a radical overhaul inside OpenAI.
MLP Networks Fit High-Frequency Functions One Oscillation at a Time
Neural networks are supposed to learn the easy stuff first. The broad strokes, the low frequencies. That’s the established story of spectral bias.
Moonshot AI seeks USD 30 billion valuation, plans USD 1‑2 billion fundraise
Moonshot AI is courting investors with a staggering $30 billion valuation. That figure, six times last year's worth, anchors its audacious pitch.
PyTorch nn.Module’s __call__ runs system setup and hooks before forward
PyTorch's nn.Module is a trap dressed as a simple container. You write your logic in the `.forward()` method. That's the bait.
AI aids meteorology and climate science without replacing experts
The hype around artificial intelligence often paints a future where algorithms replace human expertise wholesale.
SafeGene Introduces Reusable Safety-Adapter for Cross-Task Model Families
Safety in AI models is a sticker you peel off and reapply every single time you change anything. That’s the industry norm.
FAIR-Calib Introduces Two-Stage PTQ Framework for Diffusion LLM Quantization
Quantizing a diffusion language model is like tightening a screw on a running engine. Standard methods crush the delicate parts.
Elmes* Automates Fine-Grained Rubric Building for LLMs in Niche Education
Standardized tests can’t capture what makes a great tutor, or a failing one. In niche education, where every subject, grade, and task demands its own...