Sign in

Yoav Goldberg

@yoavgo.bsky.social
6.7K followers 934 following 1.4K posts
PostsRepliesMedia
Yoav Goldberg @yoavgo.bsky.social · 22/08/2026
הנה למשל דוגמה לזבל. "דקולוניזציה ותעשיית הטרור", מאמר שקורא לדקולוניזציה של "תעשיית הטרור" שמרכזה בצפון הגלובלי וארה"ב. מצד אחד המטרה האידאולוגית ברורה, מצד שני הוא עושה ככל שביכולתו לא לתת אמירה מפורשת על מה בעצם אמור לקרות. עמודים על גבי עמודים של קשקשת אקדמית www.tandfonline.com/doi/full/10....
tandfonline.com
Decolonialisation and the Terrorism Industry
Decolonising academia has gained much traction in some global north and global south countries over the last few years, resulting in initiatives such as decolonising the curricula. However, the ter...
000
Yoav Goldberg @yoavgo.bsky.social · 22/08/2026
צללתי בטעות שוב לבור הזבל שהוא מאמרים בסוציולוגיה-מונחית-אידאולוגיה. וואו כמה רמות של זבל על גבי זבל. אני רוצה להאמין שיש גם תוכן ראוי בסוציולוגיה ולא הכל ככה. אבל קשה למצוא. תמליצו לי על דברים ראויים ואולי אף מעניינים?
330
Yoav Goldberg @yoavgo.bsky.social · 17/08/2026
the bane of Academia is that it is so ruined that at the same time: a) it is forced to hide behind the idiotic non-reason of plagiarism in order to take any action; b) it has no capacity of ignoring the idiotic non-reason which is plagiarism once it is invoked. so tragic.
230
Yoav Goldberg @yoavgo.bsky.social · 16/08/2026
ארדיי מבלוסקיי ארדאי מאיקס
010
Yoav Goldberg @yoavgo.bsky.social · 30/06/2026
whats your opinion re ACL submissions that have nothing to do with NLP (beyond the relation to LLM)? lets say a challenging tool use benchmark, but where nothing is specifically about human language, language use, etc?
4100
Yoav Goldberg @yoavgo.bsky.social · 13/06/2026
which industry, what role
050
Yoav Goldberg @yoavgo.bsky.social · 04/06/2026
למה כדאי לעשות מסטר בבינה מלאכותית בימינו? כי אנחנו עובדים על סוכני AI שידעו לבצע מחקר ברמה של מסטרנט, אז עוד כמה שנים כבר לא תהיה לכם הזדמנות
1130
Yoav Goldberg @yoavgo.bsky.social · 21/05/2026
my layman remarks: technically it didnt "solve a problem", it "improved a bound". the actual underlying question, what is the optimal number, remains open. erdos was a bit arrogant and a demi god and coined "is *my* construction optimal", but that was never *the* question.
1120
Yoav Goldberg @yoavgo.bsky.social · 08/05/2026
coding agents are not compilers from English to programs. and it is not because they are non-deterministic. gist.github.com/yoavg/b2454c...
gist.github.com
Coding agents are not compilers
Coding agents are not compilers. GitHub Gist: instantly share code, notes, and snippets.
1405
Yoav Goldberg @yoavgo.bsky.social · 12/03/2026
lol how i missed bsky
200
Yoav Goldberg @yoavgo.bsky.social · 12/03/2026
i heard it is possible to scrutinize and/or criticize a technology without being delusional
6712
Yoav Goldberg @yoavgo.bsky.social · 20/12/2025
play this www.puzzlescript.net/play.html?p=...
074
Yoav Goldberg @yoavgo.bsky.social · 05/12/2025
I complain a lot about RL lately, and here we go again. The CS view of RL is wrong in how it thinks about rewards, already at the setup level. Briefly, the reward computation should be part of the agent, not part of the environment. More at length here: gist.github.com/yoavg/3eb3e7...
gist.github.com
rl-wrong-about-rewards.md
GitHub Gist: instantly share code, notes, and snippets.
2162
Yoav Goldberg @yoavgo.bsky.social · 27/11/2025
the fascinating (to me) quality of hard-core RL researchers (e.g Sutton) is the ability to have an all encompassing view of RL as the basis of intelligence, while at the same time working on super low level stuff like tabular TD algorithms, and yet strongly believe these are actually the same thing
1200
Yoav Goldberg @yoavgo.bsky.social · 17/11/2025
what's the latest-and-greatest attempt to reverse-engineer and document the inner-working of claude-code?
010
Yoav Goldberg @yoavgo.bsky.social · 16/11/2025
lets talk about "In context learning". it is clearly NOT "learning", because its ephemeral. It IS some form of generalization from examples, which is very cool. but we need a name. how do we call this skill of generalization from example?
13273
Yoav Goldberg @yoavgo.bsky.social · 08/11/2025
חשבתי שההתעלמות מסודאן היא סתם סוג-של הזנחה, אבל הנה אני קורא ספר שמתאמץ מאד להראות ש: - מה שהיה בדארפור זה לא ג'נוסייד אלא רק תוצאת לוואי של מלחמת אזרחים אכזרית - "Darfur is where the language of genocide had become an instrument" - הערבים אינם settlers באפריקה, הם הגיעו ממקומות שונים - המערב אשם
5211
Yoav Goldberg @yoavgo.bsky.social · 02/11/2025
how many of you are aware of ai-2027? how many of you read it? curious what ya'll think.
210
Yoav Goldberg @yoavgo.bsky.social · 29/10/2025
"in a future in which AI assistants are being used by world leaders and influential figures, aren't you afraid of rogue AI using this access to control the world?" well maybe, but I am more worried about near future in which these influentials act on random AI advice.
070
Yoav Goldberg @yoavgo.bsky.social · 17/10/2025
LISP code does not have significantly more parentheses than other typical languages. change my mind.
131
Yoav Goldberg @yoavgo.bsky.social · 11/09/2025
this "feature" in code editors where you type an opening character and it immediately inserts the closing one for you as well -- why?? at the very very best case, you will have to skip over this character with an arrow. so why?
4352
Yoav Goldberg @yoavgo.bsky.social · 06/09/2025
can someone explain "serverless backends" to me? it seems that they run functions on demand. but if these functions cannot access any persistent state, why not run them on the client? the only reason I see is to hide DBs/APIs tokens/secrets from the client, but is that really all there is to it?
351
Yoav Goldberg @yoavgo.bsky.social · 31/08/2025
i created this thingy yesterday and now I cannot stop watching it. yoavg.github.io/eternal/
yoavg.github.io
Eternal Struggle
99313
Yoav Goldberg @yoavgo.bsky.social · 27/08/2025
a trivia fact about this paper is that we submitted it to arxiv weeks ago, and it was hanging there in limbo for quite a while. apparently because we submitted to "AI" while they moved it to "HCI".
261
Yoav Goldberg @yoavgo.bsky.social · 27/08/2025
When reading AI reasoning text (aka CoT), we (humans) form a narrative about the underlying computation process, which we take as a transparent explanation of model behavior. But what if our narratives are wrong? We measure that and find it usually is. Now on arXiv: arxiv.org/abs/2508.16599
arxiv.org
Humans Perceive Wrong Narratives from AI Reasoning Texts
A new generation of AI models generates step-by-step reasoning text before producing an answer. This text appears to offer a human-readable window into their computation process, and is increasingly r...
48623
Yoav Goldberg @yoavgo.bsky.social · 04/07/2025
if you REALLY want to understand DL, you should start by honing your Category Theory skills, as almost everything in DL at its core can be mapped to a functor or an endofunctor.
020
Yoav Goldberg @yoavgo.bsky.social · 04/07/2025
taking it a step further, I'd say in many cases using the algebra jargon is harmful to understanding, and its better to just describe whats really going on. ie, "we add an L2 penalty term" --> want the sum of squares to be small. "project to vocab space" --> compute similarity to each vocab item.
170
Yoav Goldberg @yoavgo.bsky.social · 03/07/2025
i'll elaborate: a common computation pattern in DL happens to coincide with a known operator in linear algebra (matmul), and so we conveniently borrow linalg notation and terminology (matrices, vectors, ranks, norms). but this is just jargon. the algebric properties arent needed.
4180
Yoav Goldberg @yoavgo.bsky.social · 03/07/2025
"Modern ML is built on Linear Algebra". lol no its not.
670
Yoav Goldberg @yoavgo.bsky.social · 03/07/2025
why is "MCP" implemented as server exposing a set of endpoints, rather than as some JSON schema for defining tool descriptions and allowing these JSON files to be accessed over http? what is the purpose/benefit of the middleman server?
320
Yoav Goldberg @yoavgo.bsky.social · 22/06/2025
you know what, nah, we don't want to close it. it will be just 80% closed.
010
Yoav Goldberg @yoavgo.bsky.social · 22/06/2025
and now, we will proceed to peacefully close the strait of Hormuz. you know, for the environment.
050
Yoav Goldberg @yoavgo.bsky.social · 22/06/2025
today, during a peaceful flight over an iranian mountain, a US airplane dropped a mostly peaceful bunker buster bomb, who flew peacefully until it hit the mountain and mostly peaceful facility underneath it. there was a brief period of violent detonation on impact, then peace again.
140
Yoav Goldberg @yoavgo.bsky.social · 14/06/2025
אחד מלקחי ליל המקלטים אמש הוא שאין לי סבלנות לקרוא ספרות מקצועית, אבל לקרוא ספרות קלה זה די סבבה. מצד שני הספר שהיה לי בנייד, הוא כזה שהתחלתי לקרוא והפסקתי והיתה סיבה שהפסקתי, הוא מייגע ומעפן. בקיצור שילחו המלצות לספרים. עברית או אנגלית, אבל באנגלית ככהנ יהיה לי יותר קל להתארגן.
270
Yoav Goldberg @yoavgo.bsky.social · 14/06/2025
there are lots of "AI agents are just software around LLMs" comments recently. this is true by definition, but also misses the point. while there is no "magic" in software, there is engineering and best practices. and we dont yet have them for systems that involve LLMs.
0131
Yoav Goldberg @yoavgo.bsky.social · 08/06/2025
a good gag from the other social network:
1140
Yoav Goldberg @yoavgo.bsky.social · 08/06/2025
this "class 9" is such a cool idea for an LLM course! (from ufal.mff.cuni.cz/courses/npfl... via @zdenekkasner.bsky.social )
1121
Yoav Goldberg @yoavgo.bsky.social · 08/06/2025
i created a gist with some non default llm courses gist.github.com/yoavg/95bbc5...
gist.github.com
llm-materials-2025.md
GitHub Gist: instantly share code, notes, and snippets.
3494
Reposted by Yoav Goldberg
Grace @gracekind.net · 04/06/2025
A prompt posted by Twitter user eigenrobot
131
Yoav Goldberg @yoavgo.bsky.social · 04/06/2025
what's an "eigenprompt"?
360
Yoav Goldberg @yoavgo.bsky.social · 03/06/2025
i am looking for academic (or similarly "serious") courses or textbook-y texts on LLMs with a major component (lets say >66%) that is NOT about the technicalities of Transformers, optimization, RL, tokenization, etc, but instead focus more on data, eval, interpretability, etc. pointers?
115511
Yoav Goldberg @yoavgo.bsky.social · 12/05/2025
"hmm i wonder what [researcher person] is thinking about these days" /checks google scholar page] [50+ papers in 2025] nope, no signal there.
1181
Yoav Goldberg @yoavgo.bsky.social · 12/05/2025
an interesting observation following neurips's 25k submissions: since the main conferences in "AI" and adjacent fields grew so much, we de-facto dont have 2nd tier venues anymore. only "top tier" venues. and multiple of them. this is a weird situation.
2110
Yoav Goldberg @yoavgo.bsky.social · 11/05/2025
what's really interesting to me in ML these days is the prospect of creating systems* that learn declaratively, without losses, rewards or gradients. we are not there yet, but we will be. * not models
160
Yoav Goldberg @yoavgo.bsky.social · 09/05/2025
"this studies small models on toy synthetic data in a narrow domain. the findings may not transfer to realistic settings. unclear what we can learn from this.reject" "no no you dont understand. i study spherical frictionless lms in a vacuum. you know, like physicists!" "oh, i see. fantastic work!"
190
Yoav Goldberg @yoavgo.bsky.social · 08/05/2025
(niche) This is a nice paper but it seems they don't really understand what the word "recursive" means for CFGs, and hence their results are significantly weaker than they think.
280
Yoav Goldberg @yoavgo.bsky.social · 07/05/2025
btw, even though i dont fully understand what this doc is, i HIGHLY recommend reading it. makes you realize how much product decisions and 'behavior shaping' efforts go into "the AI" that you see at the end model-spec.openai.com/2025-04-11.h...
180
Yoav Goldberg @yoavgo.bsky.social · 07/05/2025
who here understands what OpenAI's "Model Spec" actually is? Its an interesting doc and I understand the sentiment, but what is it? just a wish list for desired behavior? an annotation guideline? a set of system prompts? llm as a judge prompts? a mix? model-spec.openai.com/2025-04-11.h...
140
Yoav Goldberg @yoavgo.bsky.social · 05/05/2025
watching a youtube video and got reminded of a nostalgic assembly trick: replacing PUSH PUSH CALL PUSH PUSH PUSH CALL PUSH CALL with PUSH PUSH PUSH PUSH PUSH PUSH CALL CALL CALL to shave a few bytes thanks to better compressibility. those were the days!
050
Yoav Goldberg @yoavgo.bsky.social · 05/05/2025
I learned that @emilymbender.bsky.social suggests replacing the term "AI" with "Automation", and I think we finally agree on something in this space! I also think that "Automation Intelligence" is a nice name for describing many of these systems and abilities.
4141