QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents Paper • 2606.32034 • Published 13 days ago • 12
spellbound-eng/hy3-spellbound-simpo-r256-b10-g3-sft0p1-lr2e-6-step303-merged-NVFP4A16-model-free Text Generation • 166B • Updated Jun 3 • 3 • 2
When No Benchmark Exists: Validating Comparative LLM Safety Scoring Without Ground-Truth Labels Paper • 2605.06652 • Published May 7 • 5
Mean Mode Screaming: Mean--Variance Split Residuals for 1000-Layer Diffusion Transformers Paper • 2605.06169 • Published May 7 • 238