Evaluating Dynamical Fidelity through Predictive Structure in Physical Representations
Research
Research News
New papers, breakthroughs and academic AI research
LLM-as-a-judge validity is strongly task-dependent across physics assessment formats
Research
The Detectability Gap: Hidden Heterogeneity in Hallucination Detection Across Language Models
Research
Reconstructing Implicit Scientific Knowledge: Evaluating LLM Agents through End-to-End Reproduction of Astronomy
Research
Watch the Model Think: On-Policy Extraction of Activation Steering Vectors
Research
GenMem: Generative Symbolic Memory for Self-Evolving Harness
Research
Position: Let's Strengthen Verifiability If We Can't Enforce Reproducibility
Research
PACT: Pairwise-Anchored Calibrated Tuning for Single-Token Typed Decisions
Research
Similarity Is Not Validity: Defending LLM Semantic Caches Against Poisoning
Research
Attention-Scoped Guidance: Training-Free Spatial Control for Image Editing
Research
MotionMaestro: Masked Tokenization for Unified Motion Generation
Research
Principled MAP estimation for inverse problems: bridging the gap between convergence and performance
Research