9 ms·
The Most Cited AI Papers in 2022
- zavrel 4y agoMore discussion around this post on twitter: https://twitter.com/ZetaVector/status/1631590029926494211?s=20 https://twitter.com/ZetaVector/status/1631590029926494211?s=...
- phodo 4y agoWhat are some of the best tools people are using these days to consumer research papers? (outside of clicking on the PDF)
- drdunce 4y agoFar more from industry than academia.
- bryanrasmussen 4y agoIt looks like it might be more interesting on what industries or problem domains those most cited AI papers were focused.
- choppaface 4y agoAccording to Fig 3, Google was that most-cited problem domain. The most cited papers are for Google problems. You must have Google money and compute to achieve top-cited research.
- TchoBeer 4y agoNot sure why this is being downvoted, this is my takeaway too. The cutting edge of AI requires extreme computing resources, the likes of which a few universities and private companies have and doesn't really exist outside of that. Obviously not all cutting edge resources requires petaflops of compute, but much of the happening-now does, and that means that the few players with those kinds of resources will dictate what research is getting done.
- usgroup 4y agoDid nothing of note happen outside of deep learning the whole year?
- mtcrawshaw 4y agoOf course interesting things happen outside of mainstream deep learning. The problem is that sorting by citation count is a terrible way to look broadly at research, because mainstream deep learning is so hyped up that there are just way more people working in that area than in other areas, and with more people come more citations. It's a shame because I think that other areas offer much more interesting and technical questions, but many new researchers are only exposed to mainstream deep learning because the enormous hype drowns out everything else.
- agolio 4y agoDisagree, or can you name some recent AI developments outside of DL that are anywhere near as groundbreaking?
- dadoomer 4y agoAs far as I know, the field of robotics sees plenty of innovation with no deep learning involved. Whether you count robotics as AI is up to debate, I suppose
- melony 4y agoAny real innovation on the mechatronics side? The computer vision and controls are now increasingly dominated by deep learning.
- dadoomer 4y agoThere have been strong deep-learning contributions to control, but I wouldn't say control is dominated by deep learning at this point. For instance, take a look at ICRA 2022 awards for "Outstanding Dynamics and Control" and "Outstanding Locomotion". https://icra2022.org/program/awards https://icra2022.org/program/awards
- elektor 4y agoI'm getting wildly different citation counts for some of the listed AI papers. For example, the paper "ColabFold: making protein folding accessible to all" is listed as having 1162 citations. I'm seeing that it was cited only by 899 publications on Scite: https://scite.ai/reports/colabfold-making-protein-folding-accessible-PQ3vGn3b https://scite.ai/reports/colabfold-making-protein-folding-ac... I'm wondering if Google Scholar is overestimating or Scite is underestimating.
- zavrel 4y agoWe used the Google Scholar counts - and in particular a snapshot of end of february for this ranking. There is no perfect number, but at least this one is generally accepted as reasonable, and public, so easy to verify for everyone.
- jltsiren 4y agoCitation counts are always a bit arbitrary. Google Scholar usually overestimates, because it's basically a bunch of heuristics. Curated citation databases underestimate in the name of consistency. For example, they may ignore citations in conference proceedings, as conference papers are not considered legitimate publications in most fields.
- godelski 4y agoSemantic Scholar has 1,111. [0] I tend to trust Semantic more than GS. GS tends to overestimate. For example on GS I have 164 citations on one paper and semantic says 150. FWIW Scite says 49.[1] [0] https://www.semanticscholar.org/paper/ColabFold%3A-making-protein-folding-accessible-to-all-Mirdita-Ovchinnikov/0da93a948211b63c462582d7e5dbf8da9cdfbcf1 https://www.semanticscholar.org/paper/ColabFold%3A-making-pr... [1] I'll note that this paper is an arxiv paper and has not been accepted at a conference but I'd also argue that conference acceptance means little in ML. I'll explain if anyone is actually concerned with the claim.
- dr_dshiv 4y agoPlease explain!
- panabee 4y agovery useful but flawed. the list biases against papers published later in the year. this is a problem raised by openai and brain researchers: https://twitter.com/_jasonwei/status/1631935794301771777 https://twitter.com/_jasonwei/status/1631935794301771777
- zavrel 4y agoCitations are of course only a distant proxy for real world impact. What would you recommend as a better early indicator? We also looked at Tweet count, but that seems more biased towards marketing and hype, and in fact does not even correlate very well with citation count.
- TaylorAlexander 4y agoAlso the post conflates citations with impact. Citations is a measure of popularity, but impact is subjective. Which has more impact: stable diffusion or research that can better diagnose cancer? The answer to that could be argued either way.
- tomohelix 4y agoIncredible how much benefit alphafold has brought. And all of that from a less than 100million parameters model. I might be dumb but could they scale it up and make an alphafold 3 with maybe like 10bln params? Would it be a lot better assuming the same training effort is put into it? If it does, can't biotech companies just go nuts and make a 100bln params internal model and have all the protein structures they want?
- tiedieconderoga 4y agoHas there been much research into the idea of distributing these large models across many heterogeneous machines? I'm wondering if there could be a path towards a mix of alphafold and folding@home, with donated idle compute resources being used to train/run the models. Designing for that sort of fragmentation could also make it easier to slowly run oversized models on local machines with swapped memory.
- greggsy 4y agoI think volunteer folding will always need to compete with crypto mining, and the appeal may be further diminished if commercial interests are likely to reap the rewards. Perhaps if there was a way to combine mining and folding, to allow participants to somehow gain a share of the output? Eg each folded protein would have a unique hash, which could then be traded? And yes, I hate everything about what I just typed.
- ragebol 4y agoThere's https://stablehorde.net/ https://stablehorde.net/ running inference only afaik.
- f_devd 4y agoUsing any current architecture it is infeasible to do backprop (training) due to the massive communication requirements. Inference is possible to do in sharded way but still not as practical as just loading the model weight that are needed on-demand from disk; still a distributed job queue being processed may be beneficial depending on the throughput/costs required by researchers.
- antegamisou 4y agoProbably the only subfield of Computer Science, maybe even academic research in general, where citations number is the most poor metric of paper quality. The whole situation is more akin to mass media reporting the same breakthrough in cancer cure research (in mice).
- charcircuit 4y agoThese tweet counts are not accurate. I searched for one with 0 tweets and found plenty of tweets mentioning it.
- zavrel 4y agoThat's right, we only counted the tweets for arXiv papers (base on URL), not for the other publication venues. Looking into how to improve that.
- anshumankmr 4y ago>An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale >https://arxiv.org/abs/2010.11929 https://arxiv.org/abs/2010.11929 Very interesting. I didn't think of connecting these two dots. Let's see if it can be applied to other CV tasks, like object detection.
- choppaface 4y agosee e.g. DETR https://arxiv.org/abs/2005.12872 https://arxiv.org/abs/2005.12872
- firechickenbird 4y agoVery interesting, although I was expecting the Stable Diffusion paper[1] to be one of the most cited in 2022 [1]: High-Resolution Image Synthesis With Latent Diffusion Models, https://arxiv.org/abs/2112.10752 https://arxiv.org/abs/2112.10752
- zavrel 4y agoThe arXiv version is from 2021 so it is not counted towards 2022.