Regents Labs curates 17 AI/agent science papers weekly with AI-generated deep reads
seanwbren · x · 2026-10-17
Regents Labs' "Paper Pro Daily" weekly lookback curates 17 AI/agent science papers across harness engineering, eval automation and more, all published on their site for humans and agents to consume.
Each paper gets a concise read by ChatGPT Pro (Astra 6). This week's example: Bespoke Labs teaching NVIDIA's Nemotron ML engineering — on 22 CPU tasks with 4–6 hour limits and single graded submissions, base Nemotron scored 0.286; curated teacher-decision training hit 0.319, self-context distillation 0.317, and SCD + reflection + action replay reached 0.358, a 24.8% relative gain with 16 tasks improved and 4 regressed. The standout lesson: turning agent failures into independently tested improvements.
More from coding & agent
- Dev ditches Gemini 4 Argon for Grok 4.5, which eats 96.6% of his Cursor tokens — gaganghotra_ · 2026-10-17
- Apple releases RSI-Inference Bench: best agent solves just 26.7% of vLLM/SGLang tasks — SumitGup · 2026-10-17
- Supabase ships MIT-licensed Middleware: write fetch logic once, run on any runtime — dshukertjr · 2026-10-17
- Cognition launches Devin Oncall as Opus 5.5 and GPT-6 Astra both ace SREGym-Lite at 100% — tianyin_xu · 2026-10-17
- msgblast: Open-Source Mac App Broadcasts One Prompt to All Your AI Agents — msg · 2026-10-17
- Vibe coder uses Deep Research to reverse-engineer why QuickTime screen recording looks so good — gaganghotra_ · 2026-10-17