Activation Flow: Manufacturing Activations for Steering
Research
Research News
New papers, breakthroughs and academic AI research
3dgs-sc: a controlled static screen-content benchmark for 3d gaussian splatting
Research
When Helpful Text Hurts: Option-Redirecting Bias in Vision-Language Models
Research
From Outcomes to Strategies: Learning Strategy Utility for Mathematical Reasoning
Research
LLM Unlearning Evaluation with TRIAGE
Research
DAAF: From Failure Localization to Editable System Assets in LLM Agents
Research
In-Token Learning for High-Fidelity Image Restoration via Diffusion Transformers
Research
Depth Laws for the Precision Floor of Trained Neural Networks: Amplification, Residual Scaling, and a Quantization-Aware Training Paradox
Research
SynDORBench: Evaluating LVLM Perceptual Robustness Under Physically Constrained Visibility Conditions
Research
The Key Handoff: Retrieval in Hybrid Language Models
Research
Separating Diagnosis from Disease Representation: Dual-View EEG Learning with Neural-Dynamics-Guided Deformation
Research
STR: Supervised Transcoder Replacement for Reducing Steering Side Effects
Research