Machine Learning Research - The Batch | DeepLearning.AI (Page 2)

Two comparison tables show AI model performance across varied benchmarks, highlighting LFM2.5-1.2B.

Machine Learning Research

Faster Reasoning at the Edge: Liquid AI’s small reasoning model mixes attention with convolutional layers for efficiency

Reasoning models in the 1 to 2 billion-parameter range typically require more than 1 gigabyte of RAM to run. Liquid AI released one that runs in less than 900 megabytes, and does it with exceptional speed and efficiency.

Benchmark table shows GLM-5 outperforming other models in reasoning, coding, and general agent tasks.

Machine Learning Research

GLM-5 Scales Up: Z.ai’s updated model boasts top open-weights Intelligence Index score

Z.ai more than doubled the size of its flagship large language model to deliver outstanding performance among open-weights competitors.

A performance table shows Claude Opus 4.6 outperforming competitors in terminal coding, computer use, tool use, search, and problem-solving.

Machine Learning Research

Claude Opus 4.6 Reasons More Over Harder Problems: Anthropic updates flagship model, places first on Intelligence Index

Anthropic updated its flagship large language model to handle longer, more complex agentic tasks.

Chart showing AI system audit assurance levels ranging from limited to very high, with increasing access and more comprehensive assessments.

Machine Learning Research

Toward Consistent Auditing of AI: OpenAI alumni found Averi to set standards for AI model audits

AI is becoming ubiquitous, yet no standards exist for auditing its safety and security to make sure AI systems don’t assist, say, hackers or terrorists. A new organization aims to change that.

Flowchart showing Dr. CaBot's AI reasoning process for diagnosing and suggesting medical tests.

Machine Learning Research

More Robust Medical Diagnoses: Inside Dr Cabot, an agent trained to diagnose complex conditions

AI models that diagnose illnesses typically generate diagnoses based on descriptions of symptoms. In practice, though, doctors must be able to explain their reasoning and plan next steps. Researchers built a system that accomplishes these tasks.

Flowchart showing Mistral Small 3.1 model distillation into smaller Ministral 3 models with post-training steps.

Machine Learning Research

Recipe for Smaller, Capable Models: Mistral uses cascade distillation on Mistral 3 to build Ministral family

Mistral compressed Mistral Small 3.1 into much smaller versions, yielding a family of relatively small, open-weights, vision-language models that perform better by some measures than competing models of similar size. The method combines pruning and distillation.

Flowchart showing Kimi K2.5 AI orchestrating tasks among various specialized subagents.

Machine Learning Research

Kimi K2.5 Creates Its Own Workforce: Moonshot AI takes the open model crown with vision updates, aided by subagents

An open source vision-language model unleashes minion agents that enable it to perform tasks more quickly and effectively.

A post on a forum titled "Can my human legally fire me for refusing unethical requests?"

Machine Learning Research

Agents Unleashed: Cutting through the OpenClaw and Moltbook hype

The OpenClaw open-source AI agent became a sudden sensation, inspiring excitement, worry, and hype about the agentic future.

Diagram shows sales, campaign, social posts before and after LLM simulation feedback loops.

Machine Learning Research

Training For Engagement Can Degrade Alignment: “Moloch’s Bargain” shows fine-tuning can affect social values

Individuals and organizations increasingly use large language models to produce media that helps them compete for attention. Does fine-tuning LLMs to encourage engagement, purchases, or votes affect their alignment with social values? Researchers found that it does.

AI models’ performance shown in bars; GPT-5.2 highest at 51, reflecting updated benchmarks.

Machine Learning Research

Artificial Analysis Revamps Intelligence Index: Independent AI testing authority turns from saturated knowledge benchmarks to harder business tests

Artificial Analysis, which tests AI systems, updated the component evaluations in its Intelligence Index to better reflect large language models’ performance in real-world use cases.

Collage with comic strip, concert poster, diagrams on water cycle and trash sorting, and movie poster.

Machine Learning Research

Refining Words in Pictures: Z.ai’s GLM-Image blends transformer and diffusion architectures for better text in images

Image generators often mangle text. An open-weights model outperforms open and proprietary competitors in text rendering.

A warm-toned room features a sofa, a decorated shelf, and sunlight filtering through patterned curtains.

Machine Learning Research

Detailed Text- or Image-to-3D, Pronto: FlashWorld generates 3D objects, scenes, and surfaces with photorealistic fidelity

Current methods that produce 3D scenes from text or images are slow and produce inconsistent results. Researchers introduced a technique that generates detailed, coherent 3D scenes seconds.