FourSight: Stable FP4 RL Post-Training via Sequence Filtering and Sparse Channel Compensation
cs.AI, cs.LG
Submitted: 2026-11-03
Updated: 2026-11-03
Terminology
Sources
- Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
- Training Verifiers to Solve Math Word Problems
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- QaRL: Rollout-Aligned Quantization-Aware RL for Fast and Stable Training under Training--Inference Mismatch
- Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference
- QuRL: Efficient Reinforcement Learning with Quantized Rollout
- Every Step Evolves: Scaling Reinforcement Learning for Trillion-Scale Thinking Model
- HiFloat4 Format for End-To-End Reinforcement Learning Post-Training of Large Language Models
- Score Centering Stabilizes Off-policy Reinforcement Learning
- FP8 Formats for Deep Learning
- Kimi K3: Open Frontier Intelligence
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- HybridFlow: A Flexible and Efficient RLHF Framework
- Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
- Optimizing Large Language Model Training Using FP4 Quantization
- Qwen3 Technical Report
- PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel
- SGLang: Efficient Execution of Structured Language Model Programs
- Diagnosing Training Inference Mismatch in LLM Reinforcement Learning via a Zero-Mismatch Reference
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents