←
Xiao2,678 posts
Xiao avatar

Xiao

@xiaobird

Associate Professor @Concordia University | PI of Health-X Lab | Medical imaging researcher & oxford comma user #neurosurgery #medicalAI #VR #ImageComputing

◷ Joined March 2010▧ 331 Photos and videos
Matthias Niessner avatar
⟳ Xiao retweeted
Matthias Niessner@MattNiessner·8h

Academia has a problem: Every author wants to use LLMs to write papers, but no reviewer wants to read AI (assisted) content. And every reviewer wants to use LLMs for reviews, but no author wants their paper judged by an AI. Authors & reviewers are the same people though...

Xiao avatar
Xiao@xiaobird·Nov 17

CVPR deadline done. The number of submissions is breaking records every year, but it seems like more and more people cannot answer questions about their work on site.

Nature ダイジェスト/編集部 avatar
⟳ Xiao retweeted
Nature ダイジェスト/編集部@NatureDigest·Nov 13

ヒト、マカク、マーモセットの大脳基底核の約2550万個の細胞を解析した、これまでで最も詳細なアトラスが作成された。運動や認知、意思決定に関わり、パーキンソン病や統合失調症、嗜癖とも関連する脳領域で、9カ国の数百人の科学者が参加した。Cellなど6誌に26本の論文。 nature.com/articles/d41586-0…

Xiao avatar
Xiao@xiaobird·Nov 14

Flood, office outage, protest/riot, Microsoft outage, flight pause. The past 7 days is a lot 🫠

PNASNews avatar
⟳ Xiao retweeted
PNASNews@PNASNews·Oct 22

A study of articles suspected of being associated with “paper mills”in the field of molecular cancer shows the papers cite other paper mill papers to inflate their metrics. These practices can amplify unreliable findings and slow scientific progress. ow.ly/1izn51008lE

Post attachment
Xiao avatar
Xiao@xiaobird·Oct 23

Congrats to our MSc trainee Erin Rainville for completing her thesis defence with an outstanding rating!

Post attachment
Xiao avatar
Xiao@xiaobird·Oct 11

Congrats to Health-X MSc trainee David Perron for his talk "Beyond Score-Based Gamification: Designing Spatiotemporal and Musical Experiences for VR Neck Rehabilitation" at ISMAR 2026 @ismarconf! 🗒️ arxiv.org/abs/2608.01688

Post attachment
Gabriela de Queiroz avatar
⟳ Xiao retweeted
Gabriela de Queiroz@gdequeiroz·Oct 1

A small request for conference season: please stop letting Claude design all your slides. We can tell. They all look the same.

Xiao avatar
Xiao@xiaobird·Oct 1

It is a wrap for MICCAI 2026! Although I couldn't be there myself this time, it was a fantastic experience for our trainees. looking forward to the next year in Auckland, New Zealand! @MICCAI_Society

Post attachment
Matthias Niessner avatar
⟳ Xiao retweeted
Matthias Niessner@MattNiessner·Sep 26

(1/n) The number of research papers has gone wild. All generated by AI? I pulled stats from CVPR/ICCV/ECCV, NeurIPS/ICML/ICLR, SIGGRAPH/SA/ToG since 2006: 119k papers + 460k arXiv preprints Accepted/yr: 968 (2006) → 19,003 (2025) → 18,976 (2026 so far) Quite surprising ->🧵

Post attachment
Xiao avatar
Xiao@xiaobird·Sep 19

Paper reviews these days don’t encourage new perspectives but “safe” papers unfortunately. Most are not looking for new problems (often application-oriented) to solve, but stay in the easy-to-review lanes.

What is the role of academic computer vision research in the age of increasingly powerful large models? Is GPT-6 Astra a step change? How can a researcher have an impact today in academia? These are the questions I ask myself as I head off to ECCV 2026, a conference I’ve attended since 1992. One of my papers this year is VIGA, a method that takes an image as input and outputs a 3D Blender scene that represents that image. This is a classical inverse-graphics task and VIGA was the first method to solve it using an agentic approach. The idea is now several years old and the first version of the paper was rejected. This delayed publication significantly. After it was accepted at ECCV, it was quickly surpassed by people using Claude Code for the same purpose. Today GPT-6 Astra blows away all previous results. But we still head off to ECCV to tell the community about our invention that is now fully out of date. The way academic work often progresses is that one reads recent papers, notices that they have limitations, comes up with a new idea, explores this, publishes it, etc. Any published paper I read today is based on ideas that are at least a year old. And those ideas were based on the literature of the time, which was also a year old. That means that any paper I see at ECCV is likely two years out of date. In AI today, two years means your work is likely irrelevant. At CVPR this summer I noticed that many authors have not gotten the message. They continue to work on “old” problems that have a long history. This history is based on assumptions about how the “vision problem” will be “solved”. The truth is that it is being solved in a very different way and many of these problems are no longer relevant. Another group of papers focuses on very niche problems where large models likely fail because of insufficient data or lack of business interest. The impactful papers were largely from industry and had long author lists and massive data+compute behind them. These papers were also out of data, describing systems that had been released months before, but at least they served to provide the community with more complete documentation and analysis of commercial systems. So what should academics do? First, we need to put aside the tools we’ve used for years and start from scratch. Every project should start by trying really hard to solve the problem with existing tools. I would like to see every paper begin with a detailed experimental analysis of how existing models perform and why they fail (if they do). This gives the kind of insight we need today. Then, assuming current models fail, the solution should provide some fundamental insight that will outlive the next release of such models. Reviewers today still focus on technical novelty. This pushes people to focus on tweaking architectures rather than clearly moving the field forward. Papers need to be judged based on their novel insight and not their novel technical contribution. This is a real shift in thinking but it focuses us on what matters - progress of the field. If we want there to be a “field” of computer vision, then it can’t become a marginal backwater, focusing on esoteric problems. If you haven’t tried using Astra (or whatever comes next) to solve your problem, then you have not done your homework. This omission should be seen as negatively as not having a previous work section. Concretely, I think papers should include a new section analogous to “Related Work” where that related work is current models and how they perform on the task. Reviewers should start asking for this and expecting authors to be able to articulate their insights about the limitations of existing large models. I'm interested in your thoughts.

Xiao avatar
Xiao@xiaobird·Sep 18

The doomsday alarms keep ringing. At the end of the day, it is not that a technology is destructive, but rather people are destructive.

Xiao avatar
Xiao@xiaobird·Sep 14

This year, HealthX-Lab is proud to present six papers in the main conference and satellite events of #MICCAI2026, covering the areas of medical AI and surgical mobile augmented reality. We look forward to the exciting discussions with the community!

Post attachment