reasoning traces
The record or history of decisions made by a model, particularly in complex reasoning tasks. Analyzing reasoning traces helps understand the model's decision-making process and can inform improvements to its reasoning capabilities.
- Antidistillation Sampling
- CoT Red-Handed: Stress Testing Chain-of-Thought Monitoring
- DISC: Dynamic Decomposition Improves LLM Inference Scaling
- Multimodal Tabular Reasoning with Privileged Structured Information
- Optimizing Anytime Reasoning via Budget Relative Policy Optimization
- ReasonFlux-PRM: Trajectory-Aware PRMs for Long Chain-of-Thought Reasoning in LLMs
- Reasoning Path Compression: Compressing Generation Trajectories for Efficient LLM Reasoning
- Reinforcement Learning Teachers of Test Time Scaling
- Think or Not? Selective Reasoning via Reinforcement Learning for Vision-Language Models
- Think-RM: Enabling Long-Horizon Reasoning in Generative Reward Models
- Thinking vs. Doing: Improving Agent Reasoning by Scaling Test-Time Interaction
- Value-Guided Search for Efficient Chain-of-Thought Reasoning
- When Thinking Drifts: Evidential Grounding for Robust Video Reasoning