Actualités
This AI Paper Introduces WEB-SHEPHERD: A Process Reward Model for Web Agents with 40K Dataset and 10× Cost Efficiency
Web navigation focuses on teaching machines how to interact with websites to perform tasks such...
This AI Paper Introduces PyVision: A Python-Centric Framework Where AI Writes Tools as It Thinks
Visual reasoning tasks challenge artificial intelligence models to interpret and process visual information using both...
This AI Paper Introduces MathCoder-VL and FigCodifier: Advancing Multimodal Mathematical Reasoning with Vision-to-Code Alignment
Multimodal mathematical reasoning enables machines to solve problems involving textual information and visual components like...
This AI Paper Introduces LLaDA-V: A Purely Diffusion-Based Multimodal Large Language Model for Visual Instruction Tuning and Multimodal Reasoning
Multimodal large language models (MLLMs) are designed to process and generate content across various modalities...
This AI Paper Introduces Group Think: A Token-Level Multi-Agent Reasoning Paradigm for Faster and Collaborative LLM Inference
A prominent area of exploration involves enabling large language models (LLMs) to function collaboratively. Multi-agent...
This AI Paper Introduces GRIT: A Method for Teaching MLLMs to Reason with Images by Interleaving Text and Visual Grounding
The core idea of Multimodal Large Language Models (MLLMs) is to create models that can...
This AI Paper Introduces Differentiable MCMC Layers: A New AI Framework for Learning with Inexact Combinatorial Solvers in Neural Networks
Neural networks have long been powerful tools for handling complex data-driven tasks. Still, they often...
This AI Paper from Microsoft Introduces WINA: A Training-Free Sparse Activation Framework for Efficient Large Language Model Inference
Large language models (LLMs), with billions of parameters, power many AI-driven services across industries. However...
This AI paper from DeepSeek-AI Explores How DeepSeek-V3 Delivers High-Performance Language Modeling by Minimizing Hardware Overhead and Maximizing Computational Efficiency
The growth in developing and deploying large language models (LLMs) is closely tied to architectural...
Thinking-Free Policy Initialization Makes Distilled Reasoning Models More Effective and Efficient Reasoners
arXiv:2509.26226v2 Announce Type: replace-cross Abstract: Reinforcement Learning with Verifiable Reward (RLVR) effectively solves complex tasks...
Thinking with Many Minds: Using Large Language Models for Multi-Perspective Problem-Solving
arXiv:2501.02348v2 Announce Type: replace Abstract: Complex problem-solving requires cognitive flexibility–the capacity to entertain multiple perspectives...
Thinking Machines Launches Tinker: A Low-Level Training API that Abstracts Distributed LLM Fine-Tuning without Hiding the Knobs
Thinking Machines has released Tinker, a Python API that lets researchers and engineers write training...
