Inference-time Alignment via Sparse Junction Steering
cs.CL, cs.AI
Submitted: 2026-01-30
Updated: 2026-11-07
Terminology
Sources
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- Constitutional AI: Harmlessness from AI Feedback
- DiffPO: Diffusion-styled Preference Optimization for Efficient Inference-Time Alignment of Large Language Models
- Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluators
- The Llama 3 Herd of Models
- AI Alignment: A Comprehensive Survey
- Alignment of Language Agents
- Attention Illuminates LLM Reasoning: The Preplan-and-Anchor Rhythm Enables Fine-Grained Policy Optimization
- Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergy
- Evolutionary Guided Decoding: Iterative Value Refinement for LLMs
- RewardBench 2: Advancing Reward Model Evaluation
- Shallow Preference Signals: Large Language Model Aligns Even Better with Truncated Data?
- Well Begun is Half Done: Low-resource Preference Alignment by Weak-to-Strong Decoding
- Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning
- Qwen3 Technical Report
- Inference-time Alignment in Continuous Space
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data