I work on auctions, fair allocation, and social choice. My son and daughter non-cooperatively curate my remaining time and interests.
Joined June 2017
- Tweets 3,713
- Following 2,429
- Followers 480
- Likes 8,816
261 Photos and videos
Pinned Tweet
30 Dec 2025
״יש לי מאמר קטן שרק אתמול בא לעולם...״
שרשור
arxiv.org/pdf/2512.22051
1
4
1,041
May 18
Getting citations right:
(1) There’s no central repository of bibliographic data. Google Scholar is terrible, I used it in my first ever paper, and got an angry email from a Professor. Apparently the GS record scanned the front page of his paper and added the editors in,
25
6
103
175,116
May 20
I deleted (7) which talked about reviewers / other scholars pointing out relevant citations because indeed it doesn't relate to my main point of possible errors when formatting citations, it's just another aspect where you don't fully control your framing, and should accommodate
1
2
63
May 20
reviewers' feedback. To emphasize, you DO read the paper you're request to include, and judge if it's relevant, and I made the call before to not include such papers.
42
May 19
במהלך משחק השחמט* מכריזה בת ה(עוד מעט) ארבע: ועכשיו - נישוקים!
כמו קפבלנקה בנצחונו על טרטקובר, היא מביאה את שני הסוסים מקצוות הלוח להיות אחד לצד השני, ועושים הפוגה במשחק כדי שיוכלו לתת זה לזה נשיקות על הפה.
*היא קוראת לזה משחק ההמהמהמ כי החלק הכיף זה שהכלים אוכלים אחד את השני
1
4
369
May 19
אל מול מה שאני סידרתי, או מלמצוא את הקואורדינטה של משבצת כשילוב של אות ומספר. כשאני מסביר לה חוק, למשל שהמשחק נגמר כשאני אוכל את המלך (ניסיתי להסביר שכשיש איום על המלך היא חייבת לזוז או להגן, אבל היא הלכה off-path, אז מה לעשות...), היא אומרת ״אוקיי אבא...״ בסלחנות ובחיוך, וייב של
1
1
154
May 19
״זה נראה שאתה סתם ממציא כל מיני חוקים, אבל אם ככה כיף לך לשחק, אז נעשה לפי החוקים שאתה רוצה״
83
May 18
then you’re expected to introduce the model in full in your paper, and the results should stand up for themselves. Correct citations have nothing to do with that.
So now to #arxivgate, it seems hallucinated citations are more of a 'tell' that the paper is 'sus', but of course,
1
10
4,626
May 18
Goodhart's Law, people would just learn to fix it and dodge the scrutiny. My point in this long form description is that there are many pitfalls even without AI, and the arXiv mega-ban is bad policy, prone to false positives and honest mistakes.
5
17
4,231
May 16
I'm anti-qualified to give advice on academic-path-maxxing but my path so far was something like:
1. Work at a tech company, do the Andrew Ng Coursera ML course, Andrej Karpathy's computer vision course ('the year of the AndreX')
2. Quit my job, travel, think about game theory
Empirically, I am not sure if this statement is true. The data seems to show that if you go to Harvard econ PhD and produce literally nothing for 5 years, you'll still have better industry job options than someone academically productive from a t20 school. Unfortunate. But...
1
11
1,703
May 16
Generally I think there's a passion/status split in Academia, and it's not clear which is the right path. I stopped blogging at some point because it was too time-consuming and not precise enough to feel meaningful anymore. I ~envy people who did Academia 'right' signaling-wise.
1
3
127
May 16
By pure technical dominance, people 'in the path' often outperform outsiders, especially if they find their passion while going at it. But the Nicholas Decker path (I think that's where all the discourse started?) is golden IMO. People are just mean to him for no reason.
2
2
164