#training
- Mean Mode Screaming: Training Pathology Fix Enables 1000-Layer Diffusion Transformers research
- Prime Intellect Releases prime-rl v0.6.0 for Agentic RL on Trillion-Parameter MoE Models Prime Intellect research
- DOPD: Dual On-Policy Distillation with Advantage-Aware Token Routing research
- ByteDance Seed paper studies how high-quality domain data should repeat when scaling LLMs ByteDance research
- SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD SLAI research
- Model Spec Midtraining: How Normative Self-Knowledge Improves Alignment Generalization Anthropic research
- Quantized Reasoning Models Think They Need to Think Longer, but They Do Not Meta research
- Hidden Decoding at Scale: A New Axis for LLM Capacity Without Backbone Growth research
- TrOPD: Trust-Region On-Policy Distillation Stabilizes LLM Training When Teacher-Student Gap Is Large Samsung Research research
- FORT-Searcher: Shortcut-Resistant Training Data Framework for Deep Search Agents research
- DomainShuttle: Subject-Driven Text-to-Video Across In-Domain and Cross-Domain Scenarios research
- QUBRIC: Co-Designing Queries and Rubrics Extends RLVR to Open-Ended Reasoning Domains research
- ZPPO: Teacher-in-Prompts Knowledge Distillation Outperforms Gradient Methods for Small Reasoners NVIDIA research
- Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements National University of Singapore (Zhi Zheng, Wee Sun Lee et al.) research
- Musk says xAI's 2-trillion-parameter Grok 4.6 has finished initial training xAI industry