The Value of Thinking: Learning When to Reason in Hybrid Decision Models
cs.AI
Submitted: 2026-11-12
Updated: 2026-11-12
Code: https://github.com/zhengxuyu/HDM
Terminology
Sources
- L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning
- FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance
- Training Large Language Models to Reason in a Continuous Latent Space
- Self-Route: Automatic Mode Switching via Capability Estimation for Efficient Reasoning
- Think Only When You Need with Large Hybrid-Reasoning Models
- Language Models (Mostly) Know What They Know
- AdaCoT: Pareto-Optimal Adaptive Chain-of-Thought Triggering via Reinforcement Learning
- Prolonged Reasoning Is Not All You Need: Certainty-Based Adaptive Routing for Efficient LLM/MLLM Reasoning
- s1: Simple test-time scaling
- HRBench: Benchmarking and Understanding Thinking-Mode Switch Strategies in Hybrid-Reasoning LLMs
- RouteLLM: Learning to Route LLMs with Preference Data
- Route to Reason: Adaptive Routing for LLM and Reasoning Strategy Selection
- Proximal Policy Optimization Algorithms
- Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters
- LEV: Thinking Fast and Deep with Looped Decision Models
- A Tutorial on LLM Reasoning: Relevant Methods behind ChatGPT o1
- AdaptThink: Reasoning Models Can Learn When to Think
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