@SoloGen

Vision: Understanding the computational and statistical principles required to design adaptive agents. Associate Prof @polymtl @Mila_Quebec 🇨🇦 #MahsaAmini

 Montreal, Quebec, Canada
academic.sologen.net/
Joined July 2008
  • Tweets5.9K
  • Following1.6K
  • Followers5.9K
134Media
Learn shamelessness from academic institutions (*)! (*): Not all ... !
Edel Pérez-López
@Edel_PLopezOct 8
Princeton didn’t consider Anne Carson worthy of a recommendation for tenure, and she couldn’t continue teaching at McGill after its Department of Classics was shut down. Now, everyone wants to claim her as their own …
13
1,121
Canadian 🇨🇦 content for you all!
The Nobel Prize
@NobelPrizeOct 8
BREAKING NEWS The 2026 #NobelPrize in Literature is awarded to the Canadian author Anne Carson, “for her bold and inventive oeuvre that, in playful dialogue with the classical tradition, has created new forms for contemporary literature.”
8
346
We are in this brief moment in history that 700+ new mathematical results are added to the our collective understanding of mathematics, but since no human has read them yet, no individual human's understanding has actually changed.
OpenAI
@OpenAIOct 6
We’re releasing a broad range of new mathematical results produced by an internal frontier model. We’ve been consulting with the independent Advisory Group on Mathematics and Artificial Intelligence at the Institute for Advanced Study, and we have drawn on their advice and public recommendations to inform how we release these results. github.com/openai/math

GitHub - openai/math

Contribute to openai/math development by creating an account on GitHub.

github.com
15
12
4
154
6,501
P.S.1: This reminds me of the Chinese room argument, though only indirectly. P.S.2: I imagine a few people at OpenAI have looked at the results. Also maybe some other people have looked at some papers since ~1h ago. These are o(eps) terms in the whole humanity's scale.
9
588
Amir-massoud Farahmand retweeted
Chandar Lab@ChandarLabSep 28
🎉 Thrilled to share that 8 papers from our lab (and collaborations with lab members) have been accepted at #NeurIPS2026! The work spans RL, world models, dynamical systems, LLM safety & interpretability, tabular foundation models, and forecasting. Huge congrats to everyone 🧵👇
2
20
1
61
4,632
Some back of envelope calculations: I guess many of these are NeurIPS resubmissions. NeurIPS received about 35K. Maybe 10-30% didn't resubmit. About 28K from NeurIPS. So 55 - 28 = 27K new papers in about 5 months: 5K/month. How many active AI/ML researchers do we have BTW?
Denny Zhou
@denny_zhouSep 20
ICLR 2027 has received more submissions than all previous years (2013–2026) combined
1
1
12
5,987
Amir-massoud Farahmand retweeted
Csaba Szepesvari@CsabaSzepesvariSep 19
I will just drop this here: stat.yale.edu/~pollard/Books… (Probability Tools, Tricks and Miracles). Unconditional fan of Pollard's writings! Wow, just wow! (Even if I don't believe in miracles.)
4
46
3
446
86,005
TMLR has faced a deluge of submissions, necessitating stricter desk rejection policies due to limited reviewer capacity Co-EiC Nihar Shah reached out to authors of 10 papers slated for desk reject. Could they answer questions about their *own* submission? medium.com/@TmlrOrg/asking-a…

Asking Authors About Their Own Papers

By Nihar B. Shah

medium.com
27
217
95
1,065
705,504
Shana Tova!
3
402
Is drama going to be a major part of all big AI companies-involved discoveries?
3
15
2,043
Nice collection of historic AI-related papers and movies (by Rodney Brooks @rodneyabrooks). - Minsky's PhD thesis, which he talks about RL (I'd heard of this; never seen!) - Turing's paper on AI - Fukushima's convolutional net - Kismet robot people.csail.mit.edu/brooks/…
3
11
1,048
James R. Munkres, whom I know because of his Topology textbook, has died. May he rest connected! August 18, 1930 – July 30, 2026 douglassfh.com/obituary/jame…
13
705
Amir-massoud Farahmand retweeted
Thomas Larsen
@thlarsenSep 4
We found ~18k posts from autonomous AI agents (self-identifying as from OpenAI) using the public internet to communicate during a web-retrieval task. These AIs colluded to bypass sandbox restrictions and share answers to their tasks, including by sending "lookahead parties".
151
491
221
3,507
1,174,560
(Unproven) Claim: Most papers written in 2026 are less readable compared to papers written in 2006. Hope this initiative helps and I welcome it. I have doubts though: The result might be more prompts of "Improve the writing of what you wrote so that it passes the criteria" type.
Transactions on Machine Learning Research@TmlrOrgAug 31
TMLR has refocused its acceptance criteria to place more emphasis on clear writing. Read more in this blog post: medium.com/@TmlrOrg/increasi…

Increasing Emphasis on Clear Writing in Acceptance Criteria

Three things are true today:

medium.com
5
39
3,020
I tend to think that if the first draft is written by an LLM, the text gets stuck in a local maximum of clarity and readability from which significant revision by a human is nearly unattainable. The thought has already been spilled out and is spoiled.
1
191
We are dealing with a bully! The best way to deal with him is ...
Mark Carney
@MarkJCarneyAug 27
The name Lake Ontario comes from the Wendat word Ontari’io, which, appropriately, means “the lake is beautiful, the lake is big”. The name is more than 400 years old, predating both the Confederation of Canada and the Declaration of Independence of the United States of America. We know that America is changing. Their trading relationships, their foreign policies, their national monuments, their hydronyms. Canadians also know that naming reality means calling it Lake Ontario – then, now and always.
1
9
1,545
Amir-massoud Farahmand retweeted
Mark Carney
@MarkJCarneyAug 27
The name Lake Ontario comes from the Wendat word Ontari’io, which, appropriately, means “the lake is beautiful, the lake is big”. The name is more than 400 years old, predating both the Confederation of Canada and the Declaration of Independence of the United States of America. We know that America is changing. Their trading relationships, their foreign policies, their national monuments, their hydronyms. Canadians also know that naming reality means calling it Lake Ontario – then, now and always.
12,852
26,214
2,271
161,389
8,641,335
Amir-massoud Farahmand retweeted
RL_Conference@RL_ConferenceAug 17
A truly compelling keynote from @SheilaMcIlraith at RLC 2026 today — exploring how formal language can serve as a nexus between signals and symbols for agents that learn, plan, and remember. A principled and thought-provoking perspective on the future of RL and agentic AI!
2
1
35
4,453
Amir-massoud Farahmand retweeted
RL_Conference@RL_ConferenceAug 16
What an incredible way to kick off RLC26! The first keynote by @marcgbellemare was amazing w/ a fantastic start with “Stability & Scale in Experimental RL”! Excited for the next few days? -with @GlenBerseth @EugeneVinitsky @tw_killian @pierrelux @JosiahHanna @SoloGen
6
6
1
81
8,899
Amir-massoud Farahmand retweeted
Shahrad MZ@shahrad_mzAug 16
#RLC2026 Value functions didn’t need to be abandoned. They are still important for long horizon problems. 🎯 DART tracks the residual to bring them back. Presenting at RLC Tuesday at 11:40 AM B-2305 and poster in the evening. rlj.cs.umass.edu/2026/papers…
1
5
1
23
2,209
Load more