Ministral 3

The Confidence Dichotomy: Analyzing and Mitigating Miscalibration in Tool-Use Agents

ArenaRL: Scaling RL for Open-Ended Agents via Tournament-based Relative Ranking

ShowUI-$π$π: Flow-based Generative Models as GUI Dexterous Hands

Learning Latent Action World Models In The Wild

Dr. Zero: Self-Evolving Search Agents without Training Data

MHLA: Restoring Expressivity of Linear Attention via Token-Level Multi-Head

GlimpRouter: Efficient Collaborative Inference by Glimpsing One Token of Thoughts

X-Coder: Advancing Competitive Programming with Fully Synthetic Tasks, Solutions, and Tests

PaCoRe: Learning to Scale Test-Time Compute with Parallel Coordinated Reasoning

BabyVision: Visual Reasoning Beyond Language

Watching, Reasoning, and Searching: A Video Deep Research Benchmark on Open Web for Agentic Video Reasoning

Conditional Memory via Scalable Lookup:A New Axis of Sparsity for Large Language Models

EnvScaler: Scaling Tool-Interactive Environments for LLM Agent via Programmatic Synthesis

Chaining the Evidence: Robust Reinforcement Learning for Deep Search Agents with Citation-Aware Rubric Rewards

CaricatureGS: Exaggerating 3D Gaussian Splatting Faces With Gaussian Curvature

The Molecular Structure of Thought: Mapping the Topology of Long Chain-of-Thought Reasoning

MMFormalizer: Multimodal Autoformalization in the Wild

Thinking with Map: Reinforced Parallel Map-Augmented Agent for Geolocalization

Breaking the Sorting Barrier for Directed Single-Source Shortest Paths

GR-Dexter Technical Report

VideoAuto-R1: Video Auto Reasoning via Thinking Once, Answering Twice

RelayLLM: Efficient Reasoning via Collaborative Decoding

Token-Level LLM Collaboration via FusionRoute

RL-AWB: Deep Reinforcement Learning for Auto White Balance Correction in Low-Light Night-time Scenes

Learnable Multipliers: Freeing the Scale of Language Model Matrix Layers

GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization

MemRL: Self-Evolving Agents via Runtime Reinforcement Learning on Episodic Memory

From Failure to Mastery: Generating Hard Samples for Tool-use Agents

Choreographing a World of Dynamic Objects

Ministral 3

The Confidence Dichotomy: Analyzing and Mitigating Miscalibration in Tool-Use Agents

ArenaRL: Scaling RL for Open-Ended Agents via Tournament-based Relative Ranking

ShowUI-$π$π: Flow-based Generative Models as GUI Dexterous Hands

Learning Latent Action World Models In The Wild

Dr. Zero: Self-Evolving Search Agents without Training Data

MHLA: Restoring Expressivity of Linear Attention via Token-Level Multi-Head

GlimpRouter: Efficient Collaborative Inference by Glimpsing One Token of Thoughts

X-Coder: Advancing Competitive Programming with Fully Synthetic Tasks, Solutions, and Tests

PaCoRe: Learning to Scale Test-Time Compute with Parallel Coordinated Reasoning

BabyVision: Visual Reasoning Beyond Language

Watching, Reasoning, and Searching: A Video Deep Research Benchmark on Open Web for Agentic Video Reasoning

Conditional Memory via Scalable Lookup:A New Axis of Sparsity for Large Language Models

EnvScaler: Scaling Tool-Interactive Environments for LLM Agent via Programmatic Synthesis

Chaining the Evidence: Robust Reinforcement Learning for Deep Search Agents with Citation-Aware Rubric Rewards

CaricatureGS: Exaggerating 3D Gaussian Splatting Faces With Gaussian Curvature

The Molecular Structure of Thought: Mapping the Topology of Long Chain-of-Thought Reasoning

MMFormalizer: Multimodal Autoformalization in the Wild

Thinking with Map: Reinforced Parallel Map-Augmented Agent for Geolocalization

Breaking the Sorting Barrier for Directed Single-Source Shortest Paths

GR-Dexter Technical Report

VideoAuto-R1: Video Auto Reasoning via Thinking Once, Answering Twice

RelayLLM: Efficient Reasoning via Collaborative Decoding

Token-Level LLM Collaboration via FusionRoute

RL-AWB: Deep Reinforcement Learning for Auto White Balance Correction in Low-Light Night-time Scenes

Learnable Multipliers: Freeing the Scale of Language Model Matrix Layers

GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization

MemRL: Self-Evolving Agents via Runtime Reinforcement Learning on Episodic Memory

From Failure to Mastery: Generating Hard Samples for Tool-use Agents

Choreographing a World of Dynamic Objects