Machine Learning Research

414 Posts

Diagram of Coconut, a method training LLMs to process thought chains as vectors, comparing it to Chain-of-Thought (CoT).
Machine Learning Research

Reasoning in Vectors, Not Text: Meta introduces Chain of Continuous Thought (Coconut) to improve next-token prediction

Although large language models can improve their performance by generating a chain of thought (CoT) — intermediate text tokens that break down the process of responding to a prompt into a series of steps.
A person typing a prompt in an AI-powered mobile app with a button to improve the input.
Machine Learning Research

Mobile Apps to Order: Replit’s agent-powered mobile app expands to full app development

Replit, an AI-driven integrated development environment, updated its mobile app to generate further mobile apps to order.
AI model comparison on reasoning and test-time compute across math, science, and coding benchmarks.
Machine Learning Research

Grok 3 Scales Up: Grok 3, xAI’s new model family, improves on its predecessors, adds reasoning

xAI’s new model family suggests that devoting more computation to training remains a viable path to building more capable AI.
Diagram showing GPT-4o with and without search, highlighting task execution success and failure.
Machine Learning Research

Tree Search for Web Agents: How tree search improves AI agents’ ability to browse the web and complete tasks

Browsing the web to achieve a specific goal can be challenging for agents based on large language models and even for vision-language models that can process onscreen images of a browser.
AI model leaderboard comparing performance across tasks like math, vision, and document analysis.
Machine Learning Research

Alibaba’s Answer to DeepSeek: Alibaba debuts Qwen2.5-VL, a powerful family of open vision-language models

While Hangzhou’s DeepSeek flexed its muscles, Chinese tech giant Alibaba vied for the spotlight with new open vision-language models.
ChatGPT interface drafting a research report on retail trends, including AI, e-commerce, and inflation.
Machine Learning Research

Agents Go Deep: OpenAI’s Deep Research agent generates detailed reports by analyzing web sources

OpenAI introduced a state-of-the-art agent that produces research reports by scouring the web and reasoning over what it finds.
Diagram illustrating Moshi’s use of an LLM to process user audio input, inner monologue, and output.
Machine Learning Research

Okay, But Please Don’t Stop Talking: Moshi, an open alternative to OpenAI’s Realtime API for Speech

Even cutting-edge, end-to-end, speech-to-speech systems like ChatGPT’s Advanced Voice Mode tend to get interrupted by interjections like “I see” and “uh-huh” that keep human conversations going. Researchers built an open alternative that’s designed to go with the flow of overlapping speech.
Line charts showing performance improvements in math and science with 2.0 Flash Thinking models.
Machine Learning Research

Gemini Thinks Faster: Google’s Gemini 2.0 Flash Thinking advances in reasoning, outperforms DeepSeek-R1

Google updated the December-vintage reasoning model Gemini 2.0 Flash Thinking and other Flash models, gaining ground on OpenAI o1 and DeepSeek-R1.
Flowchart illustrating the automation of opening, editing, and saving a Word document using PyAutoGUI.
Machine Learning Research

Training for Computer Use: UI-TARS shows strong computer use capabilities in benchmarks

As Anthropic, Google, OpenAI, and others roll out agents that are capable of computer use, new work shows how underlying models can be trained to do this.
Bar chart animation showing accuracy improvements in AIME 2024 competition math models.
Machine Learning Research

Reasoning in High Gear: o3-mini, a faster, more affordable reasoning model for coding, math, and science

OpenAI introduced a successor to its o1 models that’s faster, less expensive, and especially strong in coding, math, and science.
Bar chart comparing active vs. random sampling effects on length, diversity, and toxicity after fine-tuning.
Machine Learning Research

Fine-Tuning Fine Points: Active inheritance, a smarter way to fine-tune models on synthetic data

The practice of fine-tuning models on synthetic data is becoming well established. But synthetic training data, even if it represents the training task well, may include characteristics like toxicity that impart unwelcome properties in the trained model’s output...
AI assistant processes ‘Find me a family-friendly campsite’ and suggests options.
Machine Learning Research

Computer Use Gains Momentum: OpenAI’s Operator automates online tasks with a new AI agent

OpenAI introduced an AI agent that performs simple web tasks on a user’s behalf.
Diagram of a reinforcement learning system for training LLMs, showing data and weight flow processes.
Machine Learning Research

Reinforcement Learning Heats Up: How DeepSeek-R1 and Kimi k1.5 use reinforcement learning to improve reasoning

Reinforcement learning is emerging as an avenue for building large language models with advanced reasoning capabilities.
Workflow for inverse design using deep learning to predict S-parameters and radiation in structures.
Machine Learning Research

Generated Chip Designs Work in Mysterious Ways: Researchers used deep learning and an evolutionary algorithm to design chips in minutes

Designing integrated circuits typically requires years of human expertise. Recent work set AI to the task with surprising results.
Bar chart comparing accuracy and percentile scores of DeepSeek models and OpenAI models across benchmarks.
Machine Learning Research

DeepSeek Sharpens Its Reasoning: DeepSeek-R1, an affordable rival to OpenAI’s o1

A new open model rivals OpenAI’s o1, and it’s free to use or modify.
Load More

Subscribe to The Batch

Stay updated with weekly AI News and Insights delivered to your inbox