Analytics DeepLearning.AI - The Batch | DeepLearning.AI (Page 14)

AI co-scientist workflow diagram showing a research goal assigned to specialized AI agents for hypothesis testing and ranking

Machine Learning Research

Science Research Proposals Made to Order: AI Co-Scientist, an agent that generates research hypotheses, aiding drug discovery

An AI agent synthesizes novel scientific research hypotheses. It's already making an impact in biomedicine.

AYA Vision architecture diagram showing vision encoder, multimodal merging, and LLM backbone for image processing

Machine Learning Research

Equally Fluent in Many Languages: Cohere’s Aya Vision beats multilingual rivals in text & image understanding

Multilingual AI models often suffer uneven performance across languages, especially in multimodal tasks. A pair of lean models counters this trend with consistent understanding of text and images across major languages.

A man pastes “AI GENERATED” posters on a graffiti-covered wall in an urban alley, suggesting a street art or guerrilla marketing act.

Data Points

ERNIE checks competitors with low prices: AI2’s OLMo2 32B may be the top fully open model

Google’s two new Gemini vision-language-action robotics models. Cohere’s Command A, another lightweight LMM. New China regulations require mandatory labels for AI content. Monitoring reasoning models for reward hacking or unwanted behavior.

A therapy session in a modern office where a patient lies on a couch talking to an AI-powered computer therapist.

Data Points

AI giants’ U.S. policy proposals: Gemma 3 beats bigger open weight rivals

OpenAI’s new SDK and APIs for agentic workflows. Olympic Coder, two powerful open coding models. Alibaba applies RL to emotion detection. GPT-4.5 and Claude Sonnet 3.7 top a new agent leaderboard.

Illustration of a programmer at a computer displaying PyTorch code, while a smiling colleague gives a thumbs-up in approval.

Letters

Learn the Language of Software: AI won’t kill programming. There has never been a better time to start coding.

Some people today are discouraging others from learning programming on the grounds AI will automate it.

The Batch Newsletter

DeepSeek-R1 Uncensored, QwQ-32B Puts Reasoning in Smaller Model, Phi-4-multimodal Takes Spoken Input, Training AI May Not Be Fair Use

The Batch AI News and Insights: Some people today are discouraging others from learning programming on the grounds AI will automate it.

AI model performance benchmark comparing R1 1776 and DeepSeek-R1 across MMLU, DROP, MATH-500, and AIME 2024 tests.

Tech & Society

DeepSeek-R1 Uncensored: Perplexity launches uncensored version of DeepSeek-R1

Large language models built by developers in China may, in some applications, be less useful outside that country because they avoid topics its government deems politically sensitive. A developer fine-tuned DeepSeek-R1 to widen its scope without degrading its overall performance.

Gavel striking a neural network, symbolizing legal decisions impacting AI and machine learning technologies.

Tech & Society

Judge Upholds Copyright in AI Training Case: U.S. court rejects fair use defense in Thomson Reuters AI lawsuit

A United States court delivered a major ruling that begins to answer the question whether, and under what conditions, training an AI system on copyrighted material is considered fair use that doesn’t require permission.

Phi-4 Mini multimodal architecture integrating vision, audio, and text with token merging and LoRA-adapted weights for AI processing.

Machine Learning Research

Microsoft Tackles Voice-In, Text-Out: Microsoft’s Phi-4 Multimodal model can process text, images, and speech simultaneously

Microsoft debuted its first official large language model that responds to spoken input.

QwQ-32B vs DeepSeek-R1 AI model performance benchmark across AIME24, LiveCodeBench, LiveBench, IFEval, and BFCL datasets.

Machine Learning Research

Compact Reasoning: QwQ-32B challenges DeepSeek-R1 and other larger reasoning models

Most models that have learned to reason via reinforcement learning were huge models. A much smaller model now competes with them.

Futuristic nightclub with neon lights, a dancing crowd, and a supercomputer DJ booth glowing amid fog and lasers.

Data Points

EAGLE-3 speeds up language models: And the 2024 Turing Award goes to…

Music and lyrics in one diffusion model. Manus AI’s impressive demos spark excitement and backlash. OpenAI sees AGI as a gradual evolution. Google unveils its first Gemini-branded embedding models.

A man sitting side by side with his computer at a bar as if they are having a friendly conversation.

Data Points

Qwen’s mid-sized reasoning model scores big: Sesame moves through speech models’ “uncanny valley”

Cohere’s open vision models support many languages. Jamba 1.6’s two hybrid MoE models promise more speed. Anthropic overhauls its developer console for Claude Sonnet 3.7. Mistral brings its multilingual/multimedia skills to OCR.