Detent: Batch-Invariant LLM Serving and RL Rollouts at Near-Stock Speed
cs.LG, cs.DC
Submitted: 2026-11-07
Updated: 2026-11-07
Code: https://github.com/deepseek-ai/DeepGE
Terminology
Sources
- Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
- EigenAI: Deterministic Inference, Verifiable Results
- Training Verifiers to Solve Math Word Problems
- Gemma: Open Models Based on Gemini Research and Technology
- LLM-42: Enabling Determinism in LLM Inference with Verified Speculation
- The Llama 3 Herd of Models
- DiFR: Inference Verification Despite Nondeterminism
- Defeating the Training-Inference Mismatch via FP16
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Do Large Language Model Benchmarks Test Reliability?
- Qwen3 Technical Report
- Diagnosing Training Inference Mismatch in LLM Reinforcement Learning via a Zero-Mismatch Reference
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis