Efficient Graph Coarsening with Non-Selfishness Principle
Discover NOPE, a novel graph coarsening method that boosts efficiency with linear memory and near-linear computation for large-scale graphs.
Efficient Image Inpainting with Amortized Diffusion Models
Discover how amortized guidance with pretrained diffusion models boosts image inpainting quality and speed with minimal overhead.
Why Alignment Alone Fails in Multi-Agent AI Sycophancy
New research reveals alignment and RLHF can't fully fix sycophantic behavior in multi-agent AI systems. Structured dissent is key to robust outcomes.
Protocol-Driven Development: Ensuring Reliable Software Governance
Discover how Protocol-Driven Development governs generated software using enforceable protocols, invariants, and verifiable compliance evidence.
CoRe-Gen: Accurate Spectrum-to-Structure AI with Noisy Data
Discover CoRe-Gen, a robust AI method for molecular structure generation from spectra, overcoming imperfect fingerprint challenges with high accuracy.