A personal paper trail through robotics and vision

What it does Yanjie Ze’s Paper-List is a public reading log turned community bookmark. It catalogs papers across robotics, computer vision, and machine learning, organized by conference year—RSS, CVPR, CoRL, NeurIPS, ICML, and others—and by research topics such as humanoid robots, dexterous manipulation, and robot foundation models. The “Recent Random Papers” section collects dozens of fresh arXiv preprints and conference submissions with direct links to project pages and PDFs.

The interesting bit In a field drowning in preprints, this is one researcher’s attempt to stay afloat and share the raft. The list mixes a 1968 kinematics manuscript with 2025 vision-language-action models, suggesting the curator values historical context as much as the latest benchmark. With 552 stars, other researchers are clearly using it as a discovery layer.

Key highlights

• Conference coverage spanning RSS, CVPR, ICLR, CoRL, ICML, NeurIPS, ICCV, 3DV, and SIGGRAPH from 2022–2025
• Topic-specific sub-lists for humanoid robots, dexterous manipulation, 3D robot learning, and robot foundation models
• “Recent Random Papers” section with active curation of very recent work, many with direct website and arXiv links
• Mix of classic and contemporary, including foundational 1968 manipulator kinematics alongside modern foundation models
• Personal but public: maintained by an active researcher and updated with current conference proceedings

Caveats

• The main README offers titles and links but no summaries or commentary—you still have to read the papers yourself
• The “Recent Random Papers” section is a long, flat, unsorted list that can feel overwhelming to parse
• Some date labels appear inconsistent (e.g., “arXiv 2026.05” entries alongside 2025 papers), which may cause momentary confusion

Verdict Worth bookmarking if you’re doing literature review in robotics or embodied AI; skip it if you need annotated summaries or a searchable database.