From Outcome to Representation: Tracing Reasoning Mechanisms through Integrated Policy Gradient
cs.AI, cs.CL, cs.LG
Submitted: 2026-02-02
Updated: 2026-11-04
Terminology
Sources
- KV Cache Steering for Controlling Frozen LLMs
- Controllable LLM Reasoning via Sparse Autoencoder-Based Steering
- I Have Covered All the Bases Here: Interpreting Reasoning Features in Large Language Models via Sparse Autoencoders
- The Llama 3 Herd of Models
- Triggering Chain-of-Thought via Latent Feature Interventions in Large Language Models
- Measuring Mathematical Problem Solving With the MATH Dataset
- Decomposed Prompting: A Modular Approach for Solving Complex Tasks
- Do Sparse Autoencoders Identify Reasoning Features in Language Models?
- Compositional Explanations of Neurons
- The Quest for the Right Mediator: Surveying Mechanistic Interpretability Through the Lens of Causal Mediation Analysis
- GPQA: A Graduate-Level Google-Proof Q&A Benchmark
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Unlocking General Long Chain-of-Thought Reasoning Capabilities of Large Language Models via Representation Engineering
- LLaMA: Open and Efficient Foundation Language Models
- A Comprehensive Survey on Trustworthiness in Reasoning with Large Language Models
- MAmmoTH: Building Math Generalist Models through Hybrid Instruction Tuning
- A Comprehensive Survey of Reward Models: Taxonomy, Applications, Challenges, and Future
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