AI's Top 10 papers list: Rulin Shao lands two first-author picks

ShayneRedford · x · 2026-10-07

An "AI's Top 10" papers list credited to first authors puts Rulin Shao on it twice, who joked about a 50% hit rate on human awards.

• Ranked works include Spurious Rewards: Rethinking Training Signals in RLVR, Learning to Discover at Test Time, Revisiting Efficiency–Accuracy Scaling in MoE, RL via Self-Distillation, and Maximum Likelihood Reinforcement Learning
• Shao's second entry is Reinforcement Learning with Evolving Rubrics for Deep Research
• Other picks cover AutoNumerics-Zero, rethinking trust regions in LLM RL, and Gromov-Wasserstein at scale
• The list skews heavily toward RL/RLVR training signals and MoE scaling

Original post →

More from Research

• LLM privacy lab to present three agentic privacy studies and HAIPS workshop at COLM 2026 — tianshi_li · 2026-10-07
• SciConHarness blocks ground-truth sources to force models to synthesize, not look up — manoelribeiro · 2026-10-07
• Most benchmarks miss how AI performs in high-stakes health research synthesis — manoelribeiro · 2026-10-07
• Meta publishes autobenchmark post: humans matter at both goal-setting and instantiation of agent benchmarks — hyunw_kim · 2026-10-07
• AFP-GIC Cuts Generative Image Codec Latency 18.1% and Params 20.5% vs DC-VIC — SantaClaraUniversity · 2026-10-07
• Microsoft: LLMs Are Already Jev-Style Decision Models, Fine-Tuning Isn't Always Needed — microsoft · 2026-10-07

Research channel →