Ray Becker (@raybecker.bsky.social) (@raybbecker) header
Ray Becker (@raybecker.bsky.social) (@raybbecker) profile picture

4E Cognition and Computational corpus linguistics consultant. Embeddings ≠ Meaning

Wisconsin, USA
Joined October 2010
Photo by Ray Becker (@raybecker.bsky.social) (@raybbecker)Photo by Ray Becker (@raybecker.bsky.social) (@raybbecker)Photo by Ray Becker (@raybecker.bsky.social) (@raybbecker)Photo by Ray Becker (@raybecker.bsky.social) (@raybbecker)Photo by Ray Becker (@raybecker.bsky.social) (@raybbecker)Photo by Ray Becker (@raybecker.bsky.social) (@raybbecker)Photo by Ray Becker (@raybecker.bsky.social) (@raybbecker)Photo by Ray Becker (@raybecker.bsky.social) (@raybbecker)Photo by Ray Becker (@raybecker.bsky.social) (@raybbecker)Photo by Ray Becker (@raybecker.bsky.social) (@raybbecker)
Ray Becker (@raybecker.bsky.social) retweeted
During COVID, behavioral scientists studied why some people weren't worried enough and didn't comply enough. But some of those "problematic" people were pretty reasonable. We ask what behavioral scientists missed, why, and how we can do better next time: nature.com/articles/s41562-0…
3
12
70
5,202
Ray Becker (@raybecker.bsky.social) retweeted
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
25
100
175,537
Ray Becker (@raybecker.bsky.social) retweeted
We are pleased to share our Hierarchical Bayesian framework for Interoceptive Psychophysics! Implemented in rstan, we provide a complete suite of tools spanning model comparison, parameter recovery, multifactor designs, power analysis, and more! 🎯 biorxiv.org/content/10.1101/…
5
30
1
99
6,281
Ray Becker (@raybecker.bsky.social) retweeted
A Registered Report of 7,978 people in 20 countries finds that decision-makers willfully ignore inconvenient information. Learning about the negative consequences of their choices increases prosociality, especially among guilt-prone individuals. @JF_Schul nature.com/articles/s41562-0…
14
1
37
3,141
Ray Becker (@raybecker.bsky.social) retweeted
During the APS Awards Ceremony at #APS25DC in D.C., recipients voiced their concerns about federal actions that cut research funding, restrict international students from U.S. universities, and suppress diversity in higher education. #HigherEd #Academia
3
6
661
Ray Becker (@raybecker.bsky.social) retweeted
Looking for @theabcdstudy fMRI timeseries data? It's all in the 6.0 data release, thanks to @DrDamienFair and team at ABCC 🧠: docs.abcdstudy.org/latest/do…
11
1
47
3,503
Ray Becker (@raybecker.bsky.social) retweeted
We have put up all slide decks on the tutorial website: llm-guardrails-security.gith… 🥳🥳🥳 Although I was only able to deliver the tutorial remotely due to visa constraints, I was really thrilled to learn that our tutorial received a quite full room of audience for the entire 3.5 hr! We were very happy to receive comments from the audience like "I didn't even get bored or distracted listen to the whole 3 hours because it was so engaging."
🥳🥳🥳Join us at the tutorial of 𝐆𝐮𝐚𝐫𝐝𝐫𝐚𝐢𝐥𝐬 𝐚𝐧𝐝 𝐒𝐞𝐜𝐮𝐫𝐢𝐭𝐲 𝐟𝐨𝐫 𝐋𝐋𝐌𝐬: 𝐒𝐚𝐟𝐞, 𝐒𝐞𝐜𝐮𝐫𝐞, 𝐚𝐧𝐝 𝐂𝐨𝐧𝐭𝐫𝐨𝐥𝐥𝐚𝐛𝐥𝐞 𝐒𝐭𝐞𝐞𝐫𝐢𝐧𝐠 𝐨𝐟 𝐋𝐋𝐌 𝐀𝐩𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬! Time: 14:00 - 17:30 July 27 Location: Hall CT8
1
10
3
60
9,443
Ray Becker (@raybecker.bsky.social) retweeted
We’re rolling out new weekly rate limits for Claude Pro and Max in late August. We estimate they’ll apply to less than 5% of subscribers based on current usage.
559
301
726
2,832
2,332,998
Ray Becker (@raybecker.bsky.social) retweeted
An international team of researchers has proposed an overall definition of questionable research practices (QRPs) and published a comprehensive list of them—as well as ways to detect and prevent them. @nagyt #PsychScience psychologicalscience.org/pub…
5
9
989
Ray Becker (@raybecker.bsky.social) retweeted
I'm hiring at least one post-doc! We're interested in creating language models that process language more like humans than mainstream LLMs do, through architectural modifications and interpretability-style steering.
12
50
279
44,855
Ray Becker (@raybecker.bsky.social) retweeted
Call for Submissions! Behavioral Medicine: Special Issue on Terminated Grants. Submit by June 30 at 5PM ET. Learn more here: acrobat.adobe.com/id/urn:aai…
9
9
1,646
Ray Becker (@raybecker.bsky.social) retweeted
Collecting an extensive publicly available dataset on 4 inhibitory control tasks, @GrattonCaterina et al. show that more than 1000 trials/participant are necessary to reduce within-subject variability and improve the reliability of the congruency effect. nature.com/articles/s41562-0…
20
1
59
6,464
Ray Becker (@raybecker.bsky.social) retweeted
A computer scientist’s perspective on vibe coding:
613
1,224
222
8,465
832,626
Ray Becker (@raybecker.bsky.social) retweeted
Cooked a new paper 😎 Full text coming along Here’s a sneak peak
30
26
6
224
14,670
Ray Becker (@raybecker.bsky.social) retweeted
There's a PhD opening in Shervin's (@neuroprinciples) new lab at TU Dresden! I think this is one of the best opportunities to work on computational cogsci/neuro. Find out more: docs.google.com/document/d/1…
2
36
80
8,531
Ray Becker (@raybecker.bsky.social) retweeted
A recent paper led by amazing Demet, in collaboration with Aslı Özyürek! @DemetOzer11 @ozyurek_a
1
27
1,108
Ray Becker (@raybecker.bsky.social) retweeted
We are heartbroken to announce that Brewers icon & Baseball Hall of Famer Bob Uecker passed away today at the age of 90
3,730
11,541
9,384
55,024
13,483,538
Ray Becker (@raybecker.bsky.social) retweeted
📢 It's published!!! We (28 scientists) propose 4 Ways to increase Effect Sizes (not only Sample Size) to enhance the Replicability of Neuroimaging Research on Individual Differences. 🧠 >> check this out: doi.org/10.1162/jocn_a_02297
3
69
1
260
24,489
Ray Becker (@raybecker.bsky.social) retweeted
Oh no 😔
Excited to announce Centaur -- the first foundation model of human cognition. Centaur can predict and simulate human behavior in any experiment expressible in natural language. You can readily download the model from @huggingface and test it yourself: huggingface.co/marcelbinz/Ll…
8
13
1
169
28,346