Steered LLM Activations are Non-Surjective
Paper • 2604.09839 • Published • 15
AI & ML interests
None defined yet.
aamixsh
submitted a
paper to Daily Papers 6 months ago
aamixsh
authored a
paper 11 months ago
aamixsh
authored 5
papers about 1 year ago
Revisiting the Hypothesis: Do pretrained Transformers Learn In-Context by Gradient Descent?
Paper • 2310.08540 • Published
Source-Free and Image-Only Unsupervised Domain Adaptation for Category Level Object Pose Estimation
Paper • 2401.10848 • Published
ICL CIPHERS: Quantifying "Learning'' in In-Context Learning via Substitution Ciphers
Paper • 2504.19395 • Published • 5
On the challenges of detecting MCI using EEG in the wild
Paper • 2501.17871 • Published
IA2: Alignment with ICL Activations Improves Supervised Fine-Tuning
Paper • 2509.22621 • Published • 8