The paper introduces AIMES, a framework for adaptive multi-value activation steering in large language models. AIMES builds layer‑specific bipolar directions for moral‑foundation values and uses intermediate‑layer vocabulary readouts as online observers to guide a controller that adjusts intervention strengths at each decoding step. Experiments across instruction‑tuned model families show that AIMES achieves depth‑dependent advantages over fixed joint steering and prompt‑based steering, with smaller activation‑space interventions and comparable response quality.
arXiv:2602.01654v2 Announce Type: replace Abstract: Steering vectors (SVs) offer a lightweight way to control large language models (LLMs) at inference time by shifting hidden activations, providing...
arXiv:2602. 03160v2 Announce Type: replace Abstract: Aligning Large Language Models (LLMs) with the diverse spectrum of human values remains a central challenge: preference-based methods often fail to capture deeper motivational principles.
arXiv:2602. 07356v2 Announce Type: replace Abstract: Aligning large language models (LLMs) with human values has become increasingly important as their influence on human behavior and decision-making expands.
arXiv:2609.06289v1 Announce Type: cross Abstract: As large language models (LLMs) are increasingly deployed in alignment-sensitive contexts, activation steering has emerged as a lightweight, inferenc...
arXiv:2606. 08454v1 Announce Type: new Abstract: Activation steering provides a lightweight inference-time mechanism for controlling large language models (LLMs) by modifying their internal activation vectors toward desired behaviors.
arXiv:2609.05800v1 Announce Type: new Abstract: Activation steering controls LLM behavior at inference time by adding learned directions to hidden states, but existing methods handle one concept at a...
The paper introduces Training-Free Task Vectors (TFTVs), a method for computing task-vector-like directions in large language models without fine‑tuning. TFTVs map activation steering vectors to rank‑one weight‑space edits using only forward‑pass statistics, enabling arithmetic operations such as learning, forgetting, and composing edits. Experiments show that TFTVs consistently amplify, suppress, and combine target behaviors while preserving general knowledge, outperforming other editing and steering baselines.
arXiv:2609.37825v1 Announce Type: cross Abstract: Assigning credit to intermediate steps remains a central challenge in training Large Language Models (LLMs) on multi-step reasoning tasks with sparse...
MetaSteer is a new method for steering large language models that learns nonlinear, context-dependent interventions applied to attention projection matrices. Unlike traditional linear, context-independent techniques, MetaSteer adapts its effects based on the input, requiring no linear concept-geometry assumption. Trained once on a pooled preference corpus, it transfers zero‑shot to unseen concepts and out‑of‑distribution contexts, matching or surpassing strong task‑specific baselines on multiple benchmarks and model families.
arXiv:2608. 16739v1 Announce Type: new Abstract: Reinforcement learning algorithms for Large Language Models (LLMs) are largely distinguished by their variance reduction strategy.
arXiv:2503. 02368v4 Announce Type: replace-cross Abstract: While guided decoding, especially value-guided methods, has emerged as a cost-effective alternative for controlling language model outputs without re-training models, its effectiveness is limited by the accuracy of the value function.
arXiv:2608.21664v1 Announce Type: new Abstract: Safe deployment of increasingly capable models will likely come to rely on latent-space monitoring as a complement to behavioral evaluations, especiall...