Damià Benet

AXA Postdoctoral Fellow at Institut de Physique du Globe de Paris

[Damià Benet @dbenet4]

@branquetaG @GeolegCAT Un tall geològic dels Pirineus per acompanyar l’explicació

[Damià Benet @dbenet4]

@JaimeObregon Creo que tu LLM no tiene en el training el D-Pot de Kumar (1998)

[Damià Benet @dbenet4]

@volcaholic1 Hi! How can I contact you? I've been trying in several ways. Let me know please!

[Damià Benet @dbenet4]

@volcaholic1 Hi! how can I contact you? I tried PM but I think didn’t get through :)

[Damià Benet @dbenet4]

@Red_Rat_2018 @JacquesPerplexe @PHChess @AndriyVolokitin Mmmmm... you know that France, Germany, and so on, are included in Western world right, sweetheart?

[Damià Benet @dbenet4]

@JacquesPerplexe @PHChess @AndriyVolokitin This double standard in the entire Western world is sickening

[Damià Benet @dbenet4]

@VBkramnik @BigTurt69 @chesscom @chessdom "Never event hint" but you posted this, raising suspicions on GM Navara.

He explains quite well in his post why your post could affect his reputation.

I think here you are picking the wrong "enemy" and making a huge mistake.

[Alex Vacca @itsalexvacca]

BREAKING: MIT just completed the first brain scan study of ChatGPT users & the results are terrifying.

Turns out, AI isn't making us more productive. It's making us cognitively bankrupt.

Here's what 4 months of data revealed:

(hint: we've been measuring productivity all wrong)

[Damià Benet @dbenet4]

@yannToskov @HansMokeNiemann Science is about describing previously *unknown* phenomena, not *new* phenomena.

Leela Zero's early h4-h5 is exactly the same: an advance to a deeper understanding of chess through a concept that was previously *unknown*.

[Damià Benet @dbenet4]

@Ursustgn Quan he començat a llegir he pensat que hi hauria un gir cap a pitjor però tot el contrari.

Fa bo de llegir aquesta piulada en un dia com avui. Gràcies per ajudar la Dolores!

[Damià Benet @dbenet4]

@AlvarAlonso @chessable A "copycat" repertoire for Black

[Damià Benet @dbenet4]

@Sr_Bigotis diga'ns on és trapella. Espero que no tinguis gaires seguidors pixapins ;)

[Damià Benet @dbenet4]

@mdancho84 - shared y-axis means they could overlap
- continuous density distribution results from passing a gaussian kernel or similar obscuring distribution
- the figure takes a lot of space (many journals wouldn’t allow it)

stop using “stop using…” for clickbait please

[Damià Benet @dbenet4]

@mestres_xavier Una pregunta. Ets conscient que t’estàs indignant perquè el teu fill no tindrà classe habitual perquè part del professorat farà vaga perquè es deixin de fer atrocitats com que Israel bombardegi escoles palestines? Et poses cremeta al melich?

[Damià Benet @dbenet4]

@LaSotana_ considereu convidar-la siusplau

[Hipatia @Hipatia74]

@Sifu_Boulder @Kormavov @Indiketa61 @Xaviersapiens @RAYMKIII1 No véis que sóis parte del problema. Y te lo digo por que muchos de los que estaban integrados ya no lo están. Os estáis cargando la cultura catalana vosotros solitos. Te lo dice una competidora de sardanas y una que peleaba por el catalán. Como lo hacéis no me gusta.

[Damià Benet @dbenet4]

@HansMokeNiemann I can't purchase it; I get the error "This event has ended". Is it cuz all tickets were sold?

[Damià Benet @dbenet4]

@IglesiasYosha Cool stuff! I think it is a bit early to introduce it as an AI “that plays like a human” with accuracies around 60% but it seems very promising! Can Oracle explain the moves’ rationale? this could help support whether reasoning is somewhat human-like

[Damià Benet @dbenet4]

@LevAronian i love the subtle message of this tweet and esp the amount of humor!

let me add that most of the AI models already have a decent level of stats, probability, etc. Models will need some machine *un*learning to match the human counterpart ;)

[Damià Benet @dbenet4]

@VBkramnik Why would they arrange it if the match was already won by Jospem though?

[Damià Benet @dbenet4]

@TDataScience it would be nice from your side to add one or two sentences when you reject a story though (especially if the author is trying for the first time and has read the guidelines). Smt like “story-telling should be improved” or “content is already widely covered” would suffice.