Transferring a weak model’s RL-induced log-ratio policy shift on a strong student’s own rollouts raises AIME accuracy more cheaply than imitating the weak teacher or running matched-step RL on the student.
Towards Scalable Automated Alignment of LLMs: A Survey
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
abstract
Alignment is the most critical step in building large language models (LLMs) that meet human needs. With the rapid development of LLMs gradually surpassing human capabilities, traditional alignment methods based on human-annotation are increasingly unable to meet the scalability demands. Therefore, there is an urgent need to explore new sources of automated alignment signals and technical approaches. In this paper, we systematically review the recently emerging methods of automated alignment, attempting to explore how to achieve effective, scalable, automated alignment once the capabilities of LLMs exceed those of humans. Specifically, we categorize existing automated alignment methods into 4 major categories based on the sources of alignment signals and discuss the current status and potential development of each category. Additionally, we explore the underlying mechanisms that enable automated alignment and discuss the essential factors that make automated alignment technologies feasible and effective from the fundamental role of alignment.
representative citing papers
Pref-CTRL trains a multi-objective value function on preferences to guide representation editing for LLM alignment, outperforming RE-Control on benchmarks with better out-of-domain generalization.
ReAlignFit combines substructure edge reconstruction and a subgraph information bottleneck to align core substructure representations, and reports improved interaction prediction and distribution-shift stability on nine benchmarks.
Qwen2.5 LLMs scale pre-training data to 18 trillion tokens and apply multistage reinforcement learning, achieving competitive performance on benchmarks with models up to 5 times larger.
citing papers explorer
-
Weak-to-Strong Generalization via Direct On-Policy Distillation
Transferring a weak model’s RL-induced log-ratio policy shift on a strong student’s own rollouts raises AIME accuracy more cheaply than imitating the weak teacher or running matched-step RL on the student.
-
Pref-CTRL: Preference Driven LLM Alignment using Representation Editing
Pref-CTRL trains a multi-objective value function on preferences to guide representation editing for LLM alignment, outperforming RE-Control on benchmarks with better out-of-domain generalization.
-
Representational Alignment with Chemical Induced Fit for Molecular Relational Learning
ReAlignFit combines substructure edge reconstruction and a subgraph information bottleneck to align core substructure representations, and reports improved interaction prediction and distribution-shift stability on nine benchmarks.
-
Qwen2.5 Technical Report
Qwen2.5 LLMs scale pre-training data to 18 trillion tokens and apply multistage reinforcement learning, achieving competitive performance on benchmarks with models up to 5 times larger.