Releasing a collection of 2,000 papers to test if an LLM agent can learn research taste
ZimingLiu11 · x · 2026-08-23
Ziming Liu released a collection of 2,000 arXiv papers manually curated since 2022 to explore if an LLM agent can learn his research taste. Student Shanbin Yu built an agent to predict 15 papers he would like recently, with manual scoring and detailed comments. This remains a preliminary case study, not a formal benchmark.
More from Research
- DeepMind alumni startup's small agent outperforms OpenAI in science — emmanuelvivier · 2026-08-23
- Does training on OBLIQ tasks bake in specific similarity notions? — antoine_chaffin · 2026-08-23
- Scanned 13,350 MCP Endpoints: Open-Source Migration Checker Report — Fearzigdotss · 2026-08-23
- Open Source Project: Minimal Educational Implementation of LLM Text Watermarking — Saad_ahmed04 · 2026-08-23
- Benchmark Report: How Much Do Quants Matter on Modern Models? — KitchenAmoeba4438 · 2026-08-23
- VideoCoCo: Using Executable Code to Fix Video Physics — jiqizhixin · 2026-08-23