Teaching Models to Teach Themselves: Reasoning at the Edge of Learnability

ATLAS: Adaptive Transfer Scaling Laws for Multilingual Pretraining, Finetuning, and Decoding the Curse of Multilinguality

iFSQ: Improving FSQ for Image Generation with 1 Line of Code

Elastic Attention: Test-time Adaptive Sparsity Ratios for Efficient Transformers

Scientific Image Synthesis: Benchmarking, Methodologies, and Downstream Utility

The Script is All You Need: An Agentic Framework for Long-Horizon Dialogue-to-Cinematic Video Generation

daVinci-Dev: Agent-native Mid-training for Software Engineering

Can LLMs Clean Up Your Mess? A Survey of Application-Ready Data Preparation with LLMs

DeepSeek-OCR 2: Visual Causal Flow

Learning to Discover at Test Time

Eliciting Harmful Capabilities by Fine-Tuning On Safeguarded Outputs

Memory-V2V: Augmenting Video-to-Video Diffusion Models with Memory

Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verification

VisGym: Diverse, Customizable, Scalable Environments for Multimodal Agents

TwinBrainVLA: Unleashing the Potential of Generalist VLMs for Embodied Tasks via Asymmetric Mixture-of-Transformers

SWE-Pruner: Self-Adaptive Context Pruning for Coding Agents

LongCat-Flash-Thinking-2601 Technical Report

Can Language Models Discover Scaling Laws?

Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning

Scaling Text-to-Image Diffusion Transformers with Representation Autoencoders

BayesianVLA: Bayesian Decomposition of Vision Language Action Models via Latent Action Queries

The Flexibility Trap: Why Arbitrary Order Limits Reasoning Potential in Diffusion Language Models

LLM-in-Sandbox Elicits General Agentic Intelligence

HERMES: KV Cache as Hierarchical Memory for Efficient Streaming Video Understanding

EvoCUA: Evolving Computer Use Agents via Learning from Scalable Synthetic Experience

HY-MT1.5 Technical Report

Scaling Laws for Code: Every Programming Language Matters

Qwen3-TTS Technical Report

Small Models, Big Results: Achieving Superior Intent Extraction through Decomposition

FinVault: Benchmarking Financial Agent Safety in Execution-Grounded Environments

Teaching Models to Teach Themselves: Reasoning at the Edge of Learnability

ATLAS: Adaptive Transfer Scaling Laws for Multilingual Pretraining, Finetuning, and Decoding the Curse of Multilinguality

iFSQ: Improving FSQ for Image Generation with 1 Line of Code

Elastic Attention: Test-time Adaptive Sparsity Ratios for Efficient Transformers

Scientific Image Synthesis: Benchmarking, Methodologies, and Downstream Utility

The Script is All You Need: An Agentic Framework for Long-Horizon Dialogue-to-Cinematic Video Generation

daVinci-Dev: Agent-native Mid-training for Software Engineering

Can LLMs Clean Up Your Mess? A Survey of Application-Ready Data Preparation with LLMs

DeepSeek-OCR 2: Visual Causal Flow

Learning to Discover at Test Time

Eliciting Harmful Capabilities by Fine-Tuning On Safeguarded Outputs

Memory-V2V: Augmenting Video-to-Video Diffusion Models with Memory

Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verification

VisGym: Diverse, Customizable, Scalable Environments for Multimodal Agents

TwinBrainVLA: Unleashing the Potential of Generalist VLMs for Embodied Tasks via Asymmetric Mixture-of-Transformers

SWE-Pruner: Self-Adaptive Context Pruning for Coding Agents

LongCat-Flash-Thinking-2601 Technical Report

Can Language Models Discover Scaling Laws?

Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning

Scaling Text-to-Image Diffusion Transformers with Representation Autoencoders

BayesianVLA: Bayesian Decomposition of Vision Language Action Models via Latent Action Queries

The Flexibility Trap: Why Arbitrary Order Limits Reasoning Potential in Diffusion Language Models

LLM-in-Sandbox Elicits General Agentic Intelligence

HERMES: KV Cache as Hierarchical Memory for Efficient Streaming Video Understanding

EvoCUA: Evolving Computer Use Agents via Learning from Scalable Synthetic Experience

HY-MT1.5 Technical Report

Scaling Laws for Code: Every Programming Language Matters

Qwen3-TTS Technical Report

Small Models, Big Results: Achieving Superior Intent Extraction through Decomposition

FinVault: Benchmarking Financial Agent Safety in Execution-Grounded Environments