ByteDance's PEAR automates search research with a 4-level evidence ladder
_reachsumit · x · 2026-09-29
ByteDance presents PEAR, an AutoResearch framework for industrial search systems. It maintains an independent hypothesis-guided research state per strategy task via a Plan-Execute-Evaluate-Update loop, and a Confidence-Gated Verifier Ladder that organizes evaluation into four fidelity levels—from cheap proxies to online validation—reserving online tests for the strongest candidates. It fixes the keep-if-better rule's tendency to mistake transient gains under non-stationary traffic for persistent improvements.
More from Research
- Meta's LSRM wins ECCV 2026 Best Paper Honorable Mention, scales 3D reconstruction with sparse attention — rsasaki0109 · 2026-09-30
- Amazon's GEB grounds entity biographies for long-video memory, hits 72% on EgoLifeQA — amazon · 2026-09-30
- NVIDIA's LongLive-Plug: Distill Once, Deploy Training-Free Across 54 Downstream Video Models — nvidia · 2026-09-30
- CrossBFM Distills a Shared Behavior Space Across Humanoid Robots in Under One GPU-Hour — Tan-Dzung Do · 2026-09-30
- AnisoWM: anisotropic representations improve planning in JEPA world models — SeoulNatlUniv · 2026-09-30
- Meituan's SAKI: Coupling-Routed Teacher Supervision Speeds Up On-Policy Distillation 4.22x — meituan · 2026-09-30