OpenBMB's MiniCPM-OCR Pairs a 1.3B Parser With a 2B Accuracy Verifier
OpenBMB released a two-model OCR system: a 1.3B parser hitting state-of-the-art accuracy and a 2B verifier that scores whether its output is correct.
20 stories
OpenBMB released a two-model OCR system: a 1.3B parser hitting state-of-the-art accuracy and a 2B verifier that scores whether its output is correct.
Alibaba's Tongyi-MAI lab released a 6B text-to-image model that renders photorealistic images in under a second on a 16GB GPU, with Apache 2.0 weights.
A self-supervised method treats multiple X-ray projections of the same patient as paired data, beating supervised and vision-language baselines on three datasets.
A Hugging Face upload called burner22/numbat claims to be a mid-pretraining checkpoint of Anthropic's next Mythos-class model, around 14T parameters.
Sakana AI launches Aquarium, an ontology-based agent platform that unifies enterprise data, workflows, and governance for operational AI deployments.
A new coding agent leaderboard swaps in Cognition's FrontierCode benchmark and tightens Q&A grading, reshuffling the top of the charts.
Idiap and Uniphore researchers show that squeezing prior conversational audio into a handful of latent tokens recovers most context-aware ASR gains while slashing prompt length.
QFT-Bench puts 39 frontier and open-source models through 75 graduate and research-level quantum field theory problems, exposing where reasoning actually breaks.
A tiny 86k-parameter causal model estimates which notes are sounding within 50 ms of audio arriving, matching or beating much larger offline systems.
A new self-training framework called OPST lets language models improve themselves using only topic names, with no human-written questions, answers, or reasoning traces.
FMER swaps diffusion for flow matching in online RL, delivering tractable entropy, mirror descent updates, and up to 7x faster training.
Mistral's first time-series foundation model family runs on CPU, hits #1 on three forecasting benchmarks, and ships under a Modified MIT license.
Anthropic and Theorem used Claude to produce machine-checked no-crash proofs for 123 open-source libraries at the assembly level, uncovering long-latent bugs.
Ant Group's Ming-3.0-tts packs top-tier speech synthesis into just 0.5B parameters, released MIT-licensed and holding its own against 3.5B+ rivals.
Anthropic now runs Claude directly on Google Cloud with day-zero feature parity, GCP billing, IAM, and commitment drawdown, no Vertex AI required.
SubQ Snap 1.1 ties d1-ultra and Elephant 9B atop the Decision Index at a score of 65, winning on hallucination control and context length.
Viggle open-sourced a two-LoRA system on MiniMax H3 that recasts actors, swaps props and changes weather in existing footage while preserving motion, camera and audio.
The revised physics research benchmark gives models a code sandbox, and frontier systems have jumped from single-digit to 80%+ accuracy.
You.com's new Inference Router pairs open-weight models like Kimi K3 and GLM 5.3 with a 99.9% uptime SLA, zero data retention, and built-in web search.
Two overlooked floating-point bugs in softmax and cross-entropy have been quietly corrupting gradients. A new paper fixes them in a few lines of JAX.