Project Greenhouse: Jimmy Lin Team Trains Fully Open Reranker from Scratch on a Handful of GPUs
_reachsumit · x · 2026-10-09
Jimmy Lin's team at Waterloo released Project Greenhouse [2610.11922], testing the thesis that fully open, sovereign agentic search models can be built with modest compute.
- Method: a two-step recipe—pre-training a decoder-only pointwise reranker from scratch on public data, then supervised fine-tuning on public datasets
- Unlike the dominant approach, no reliance on third-party open-weight backbones; end-to-end control of training
- Compute: the bulk of experiments used no more than a handful of GPUs
- Artifacts: data, code, configs, and checkpoints for the Gaggle model family, enabling transparent independent reproduction
- Significance: first milestone validating that agentic search components can be built without big-lab open weights
More from Research
- RSI-Exam lands in State of AI report: top model Opus 5.5 scores just ~0.53 — cihangxie · 2026-10-09
- KAIST's ME-World tackles multi-agent egocentric world modeling with joint denoising — kaist-ai · 2026-10-09
- Controlled study: LLMs struggle to recover latent sequential structure despite long context — illinois · 2026-10-09
- TerraVis quantifies world-grounded visual consistency failures in text-to-image models — the-aiml · 2026-10-09
- OneSearch-VL: unified multimodal deep research agent beats Qwen3-VL by 20 points — Hongyu Li · 2026-10-09
- SpaceCast-Bench: best VLM hits 58.0% on predictive spatial reasoning vs 87.2% human — zju · 2026-10-09