Prompt-boundary directional alignment enables geometry-guided search that cuts trials to 95% best utility by 39.8% on average, while concept granularity predicts remaining difficulty via directional heterogeneity.
Hypersteer: Activation steering at scale with hypernetworks
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 6verdicts
UNVERDICTED 6roles
background 1polarities
background 1representative citing papers
HyperTransport amortizes activation steering for T2I models via a hypernetwork that predicts intervention parameters from CLIP embeddings, delivering 3600-7000x speedup and matching per-concept baselines on 167 unseen concepts.
PSR models that estimate token-specific steering coefficients from activations outperform standard activation steering and compare favorably to prompting on steering benchmarks.
Activation steering produces synthetic safety-violating data that improves downstream classifiers over prompting on most tested concepts when a harmonic mean of alignment, coherence, and diversity is optimized.
Steering vectors for refusal primarily modify the OV circuit in attention, ignore most of the QK circuit, and can be sparsified to 1-10% of dimensions while retaining performance.
Palette identifies refusal directions via multi-objective search, internalizes them through lightweight adaptation, and supports on-demand multi-domain authorization via independent learning and parameter merging.
citing papers explorer
-
When Is Rank-1 Steering Cheap? Geometry, Granularity, and Budgeted Search
Prompt-boundary directional alignment enables geometry-guided search that cuts trials to 95% best utility by 39.8% on average, while concept granularity predicts remaining difficulty via directional heterogeneity.
-
HyperTransport: Amortized Conditioning of T2I Generative Models
HyperTransport amortizes activation steering for T2I models via a hypernetwork that predicts intervention parameters from CLIP embeddings, delivering 3600-7000x speedup and matching per-concept baselines on 167 unseen concepts.
-
Steer Like the LLM: Activation Steering that Mimics Prompting
PSR models that estimate token-specific steering coefficients from activations outperform standard activation steering and compare favorably to prompting on steering benchmarks.
-
Activation Steering for Synthetic Data Generation: The Role of Diversity in Downstream Safety Detection
Activation steering produces synthetic safety-violating data that improves downstream classifiers over prompting on most tested concepts when a harmonic mean of alignment, coherence, and diversity is optimized.
-
What Drives Representation Steering? A Mechanistic Case Study on Steering Refusal
Steering vectors for refusal primarily modify the OV circuit in attention, ignore most of the QK circuit, and can be sparsified to 1-10% of dimensions while retaining performance.
-
Palette: A Modular, Controllable, and Efficient Framework for On-demand Authorized Safety Alignment Relaxation in LLMs
Palette identifies refusal directions via multi-objective search, internalizes them through lightweight adaptation, and supports on-demand multi-domain authorization via independent learning and parameter merging.