The Batch | DeepLearning.AI (Page 10)

A crowded street in a bustling big city with towering buildings, traffic, and people walking around.

Data Points

Google adds Thinking Mode to Flash 2.0: OpenAI’s o1 now available in the API

Speech-to-speech models fall short on benchmarks. Backlash against misleading news summaries. Google updates its video and image models. Nvidia’s $250 palm-sized computer for AI developers.

Graph showing cross-validation accuracy vs. number of features for raw and whitened inputs.

Letters

Focus on the Future, Learn From the Past: 15 years ago, the idea of scaling up deep learning was controversial — but it was right. Keep your eyes open for such ideas in 2025.

I’m thrilled that former students and postdocs of mine won both of this year’s NeurIPS Test of Time Paper Awards.

The Batch Newsletter

Phi-4 Breaks Size Barrier, HunyuanVideo Narrows Open Source Gap, Gemini 2.0 Flash Accelerates Multimodal Modeling, LLMs Propose Research Ideas

The Batch AI News and Insights: I’m thrilled that former students and postdocs of mine won both of this year’s NeurIPS Test of Time Paper Awards.

Animation showcasing 7 key NLP topics visually expanding on the screen.

Machine Learning Research

When LLMs Propose Research Ideas: Stanford study finds AI matches human experts at writing research proposals

How do agents based on large language models compare to human experts when it comes to proposing machine learning research? Pretty well, according to one study.

Performance comparison for Gemini models across benchmarks.

Machine Learning Research

Multimodal Modeling on the Double: Google introduces Gemini 2.0 Flash, a faster, more capable AI model

Google’s Gemini 2.0 Flash, the first member of its updated Gemini family of large multimodal models, combines speed with performance that exceeds that of its earlier flagship model, Gemini 1.5 Pro, on several measures.

A GIF with scenes of a man at a café, a working robot, a ghost in a mirror, and a speeding truck.

Machine Learning Research

Open Video Gen Closes the Gap: Tencent releases HunyuanVideo, an open source model rivaling commercial video generators

The gap is narrowing between closed and open models for video generation.

Benchmark results for Phi-4, GPT, LLaMA-3.3, and Qwen 2.5 models.

Machine Learning Research

Phi-4 Beats Models Five Times Its Size: Microsoft’s Phi-4 blends synthetic and organic data to surpass larger models in math and reasoning benchmarks

Microsoft updated its smallest model family with a single, surprisingly high-performance model.

A futuristic scene of humanoid robots testing AI systems in a lab with data charts and coding screens.

Data Points

Microsoft’s Phi-4 proves AI power isn’t just about size: Projects and Canvas get ChatGPT ready for work

NotebookLM gets a plus-sized update. Meta’s Motivo model marks a return to the metaverse. New synthetic data generator makes training easier. NeurIPS 2024’s sabotage controversy.

Image of a large, modern library with a vibrant technological environment.

Data Points

Harvard University releases giant dataset of public-domain books: Google’s Gemini 2.0 Flash beats top models on benchmarks

ChatGPT’s Advanced Voice Mode can see as well as hear. GM drops robotaxi division in favor of AI-assisted driving. Ruliad’s new model offers transparent reasoning in a small package. Large Concept Models move on from token-based AI.

Cartoon showing people stuck in wet concrete, with a person saying ‘You asked for a concrete idea!’

Letters

Best Practices for AI Product Management: Generative AI is making it possible to build new kinds of applications in new ways. Here are emerging best practices for AI product managers.

AI Product Management is evolving rapidly. The growth of generative AI and AI-based developer tools has created numerous opportunities to build AI applications.

The Batch Newsletter

Amazon Nova’s Competitive Price/Performance, OpenAI o1 Pro’s High Price/Performance, Google’s Game Worlds on Tap, Factual LLMs

The Batch AI News and Insights: AI Product Management is evolving rapidly. The growth of generative AI and AI-based developer tools has created numerous opportunities to build AI applications.

Graph showing how training loss affects token prediction accuracy and hallucination elimination.

Machine Learning Research

Getting the Facts Right: A memory method that reduces hallucinations in LLMs

Large language models that remember more hallucinate less.

Latest

Google adds Thinking Mode to Flash 2.0: OpenAI’s o1 now available in the API

Focus on the Future, Learn From the Past: 15 years ago, the idea of scaling up deep learning was controversial — but it was right. Keep your eyes open for such ideas in 2025.

Phi-4 Breaks Size Barrier, HunyuanVideo Narrows Open Source Gap, Gemini 2.0 Flash Accelerates Multimodal Modeling, LLMs Propose Research Ideas

When LLMs Propose Research Ideas: Stanford study finds AI matches human experts at writing research proposals

Multimodal Modeling on the Double: Google introduces Gemini 2.0 Flash, a faster, more capable AI model

Open Video Gen Closes the Gap: Tencent releases HunyuanVideo, an open source model rivaling commercial video generators

Phi-4 Beats Models Five Times Its Size: Microsoft’s Phi-4 blends synthetic and organic data to surpass larger models in math and reasoning benchmarks

Microsoft’s Phi-4 proves AI power isn’t just about size: Projects and Canvas get ChatGPT ready for work

Harvard University releases giant dataset of public-domain books: Google’s Gemini 2.0 Flash beats top models on benchmarks

Best Practices for AI Product Management: Generative AI is making it possible to build new kinds of applications in new ways. Here are emerging best practices for AI product managers.

Amazon Nova’s Competitive Price/Performance, OpenAI o1 Pro’s High Price/Performance, Google’s Game Worlds on Tap, Factual LLMs

Getting the Facts Right: A memory method that reduces hallucinations in LLMs