6 papers
Matrix: Peer-to-Peer Multi-Agent Synthetic Data Generation Framework
Dong Wang, Yang Li, Ansong Ni +12
Synthetic data has become increasingly important for training large language models, especially when real data is scarce, expensive, or privacy-sensitive. Many such generation task…
TV2TV: A Unified Framework for Interleaved Language and Video Generation
Xiaochuang Han, Youssef Emad, Melissa Hall +15
Video generation models are rapidly advancing, but can still struggle with complex video outputs that require significant semantic branching or repeated high-level reasoning about…
NaturalReasoning: Reasoning in the Wild with 2.8M Challenging Questions
Weizhe Yuan, Jane Yu, Song Jiang +8
Scaling reasoning capabilities beyond traditional domains such as math and coding is hindered by the lack of diverse and high-quality questions. To overcome this limitation, we int…
Quagmires in SFT-RL Post-Training: When High SFT Scores Mislead and What to Use Instead
Feiyang Kang, Michael Kuchnik, Karthik Padthe +4
In post-training for reasoning Large Language Models (LLMs), the current state of practice trains LLMs in two independent stages: Supervised Fine-Tuning (SFT) and Reinforcement Lea…
NaturalThoughts: Selecting and Distilling Reasoning Traces for General Reasoning Tasks
Yang Li, Youssef Emad, Karthik Padthe +8
Recent work has shown that distilling reasoning traces from a larger teacher model via supervised finetuning outperforms reinforcement learning with the smaller student model alone…
Improving Factuality with Explicit Working Memory
Mingda Chen, Yang Li, Karthik Padthe +5
Large language models can generate factually inaccurate content, a problem known as hallucination. Recent works have built upon retrieved-augmented generation to improve factuality…