RATIO: Reasoning Analysis and Token-level Inference Optimization for Quantized Reasoning Models
cs.LG
Submitted: 2026-09-30
Updated: 2026-09-30
Code: https://github.com/steven-bao1/RATIO
Sources
- Extreme Low-Bit Inference in Reasoning Models: Failure Modes and Targeted Recovery
- SOAR: Scale Optimization for Accurate Reconstruction in NVFP4 Quantization
- Evaluating Large Language Models Trained on Code
- RaZeR: Pushing the Limits of NVFP4 Quantization with Redundant Zero Remapping
- Training Verifiers to Solve Math Word Problems
- DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
- Beyond Benchmarks: MathArena as an Evaluation Platform for Mathematics with LLMs
- GLM-5: from Vibe Coding to Agentic Engineering
- The Llama 3 Herd of Models
- Kimi K3: Open Frontier Intelligence
- Quantization Meets Reasoning: Exploring LLM Low-Bit Quantization Degradation for Mathematical Reasoning
- FOCUS: FP4 Optimization via Coupled-Relaxation and Dual-Granularity Scaling
- Qwen2.5 Technical Report
- Qwen3 Technical Report
- QuantLRM: Quantization of Large Reasoning Models via Fine-Tuning Signals
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks