Does Steering Break Your Model? A Multi-Dimensional Evaluation Suite for LLM Steering Methods
Research
Research News
New papers, breakthroughs and academic AI research
Pre-Deployment Complexity Estimation for Federated Perception Systems
Research
Verified, not generated: expert-verified AI study materials and the distribution of learning gains in a university course
Research
SanSi: A Looped Typed Decision Model for System 1.5 Thinking
Research
From Evidence to Action: How Tool-Using Agents Fail
Research
When Attention Does Not Explain the Peak: Temporal Reference vs. Forecast Output in Attention-Based Time-Series Forecasting
Research
No Transformer Beats Six Covariates: Long-Horizon Prediction of Depressive Symptoms from Childhood Essays
Research
SoloQ: Calibration-Free Quantization for Diffusion Language Models
Research
SchemaFill: Efficient LLM Tool Calling via Slot-Parallel Speculative Decoding
Research
Will the Judge Flip? Predicting Position-Sensitive LLM Judgments from Residual Stream Activations
Research
Muon Is Theoretically Wrong For Convolutions, But Empirically Effective
Research
Reading, Not Manipulating: Leveraging Router Logits for Multimodal Safety in MoE Vision-Language Models
Research