Local linearity of LLM layers enables LQR-based closed-loop activation steering with theoretical tracking guarantees.
Args: Alignment as reward-guided search
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
representative citing papers
OPPO derives token-level advantages for LLM RL via Bayesian recursion on oracle signals, recovering prior distillation methods as a special case and showing gains on math and code benchmarks.
Spectral Souping learns offline specialized policies for fine-grained preferences and merges them online using a discovered universal spectral representation for efficient LLM alignment.
DISCA converts within-country disagreement among World Values Survey personas into a bounded logit correction that reduces cultural misalignment by 10-24% on MultiTP for models 3.8B and larger across 20 countries, without any weight updates.
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.
REFORM uses reward-guided controlled decoding to generate adversarial failures and augments training data to improve reward model robustness on preference datasets.
Disentangled Safety Adapters decouple safety computations from task-optimized LLMs via lightweight adapters, yielding up to 53% better AUC on safety tasks and dynamic inference-time alignment with reduced performance trade-offs.
Preference-Paired Fine-Tuning (PFT) lets LLMs handle conflicting and dynamic individual preferences better than single-preference methods, reaching 96.6% accuracy on the new VCD dataset and 44.76% gains in user alignment with limited history.
citing papers explorer
-
Local Linearity of LLMs Enables Activation Steering via Model-Based Linear Optimal Control
Local linearity of LLM layers enables LQR-based closed-loop activation steering with theoretical tracking guarantees.
-
OPPO: Bayesian Value Recursion for Token-Level Credit Assignment in LLM Reasoning
OPPO derives token-level advantages for LLM RL via Bayesian recursion on oracle signals, recovering prior distillation methods as a special case and showing gains on math and code benchmarks.
-
Spectral Souping: A Unified Framework for Online Preference Alignment
Spectral Souping learns offline specialized policies for fine-grained preferences and merges them online using a discovered universal spectral representation for efficient LLM alignment.
-
Training-Free Cultural Alignment of Large Language Models via Persona Disagreement
DISCA converts within-country disagreement among World Values Survey personas into a bounded logit correction that reduces cultural misalignment by 10-24% on MultiTP for models 3.8B and larger across 20 countries, without any weight updates.
-
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.
-
Reward Models Can Improve Themselves: Reward-Guided Adversarial Failure Mode Discovery for Robust Reward Modeling
REFORM uses reward-guided controlled decoding to generate adversarial failures and augments training data to improve reward model robustness on preference datasets.
-
Disentangled Safety Adapters Enable Efficient Guardrails and Flexible Inference-Time Alignment
Disentangled Safety Adapters decouple safety computations from task-optimized LLMs via lightweight adapters, yielding up to 53% better AUC on safety tasks and dynamic inference-time alignment with reduced performance trade-offs.
-
Meet Dynamic Individual Preferences: Resolving Conflicting Human Value with Paired Fine-Tuning
Preference-Paired Fine-Tuning (PFT) lets LLMs handle conflicting and dynamic individual preferences better than single-preference methods, reaching 96.6% accuracy on the new VCD dataset and 44.76% gains in user alignment with limited history.