PIC

山に登ったり自転車乗ったり、超かぐや姫だったり。
気に入ったものをひたすら褒めるおじさんです。

[PIC @PIC_jpn]

せっかくなのでフレンドも募集しておきます
UID:4411675805
#エンドフィールドフレンド募集
#エンドフィールド

[Tiger @tigertwting]

ok now we are at the bird civil war!!!

[PIC @PIC_jpn]

今日もTwitterで鳥たちが争ってる

[chicken @chickenincrisis]

no fuck you

[サイバー環状線 @CyberKanjousen]

公司新来的实习生，看起来年纪好大啊

[Polymarket @Polymarket]

JUST IN: Robot dog spotted taking real dog for a walk in San Francisco.

[pigeon @pigeontwting]

fuck you

[ぐしゃぐしゃの愚者 @ggg_sya]

天理の姿を考えようの会

[JustinLLL @Justin1219666]

男生保养自己的睾丸是高认知表现，我每天都会保养，等到睾丸发光就是保养到位了，上次我晚上在山里迷路，就是靠着发光的睾丸让救援队发现了我。

[なぎこ @nagiko_795]

※世界任務「結晶と石箱の古謡」より

頑張るパイモン…！

[PIC @PIC_jpn]

日本語の全二重が増えたけど、結局Moshiベースだからこの手の問題が出てくるんだよね。
少なくとも12.5Hzに押し込む関係で、これ以上賢くはできない。

深い思考を別モデルに非同期で委譲する事自体は苦肉の策と言うほどでもなく、真っ当な解決策にも思ってる。

[pulsatingGenius @Neuralwarlock]

the main problem with current full duplex models is that almost everyone is iterating on the same moshi architecture which imo has some real bottlenecks.

in moshi every text token is locked 1:1 to an audio frame that gets spoken right away (apart from padding and special tokens). because of this rigid lockstep the native architecture cant generate a burst of text without advancing audio time. if the model generates 20 text tokens time has to move forward 20 frames so exactly 1.6 seconds of audio gets synthesized alongside. you cant pause the audio clock to let the LLM write out a block of text.

within base moshi you could in theory train it to output reasoning text or JSON tool calls while forcing the audio stream to output silence tokens. but since the clock cant be decoupled that "silent thinking" still runs at 12.5 tokens per second. a 50 token thought chain or tool call means the user sits through 4 solid seconds of awkward dead air.

the other option is synthetic data where it keeps talking while producing the JSON or the thinking. but there is no data in the wild for that so you generate a lot of it and now your data is the bottleneck. the voice will only ever be as natural as the most natural TTS you can use it will always carry the quirks of the harness you built to create it and if whoever is making that data cant think through the scenarios of dyadic conversations well, you are cooked.

so people came up with delegation system which is a pain in itself. you train the frontend model to be smart enough to hold a decent conversation and to hand off tool calling or thinking to a much larger, more capable model. this is engineering and its the only thing that has made the 2 or 3 full duplex models we have like openai gpt live possible at all.

thinking cant be done in speech. the signal to noise ratio is low and there isnt enough speech data to give the model a good enough world model in its weights. how much data is there of people doing maths out loud? even we humans dont think in speech. yes there is "thinking out loud" but speech is not the medium humans think in. thinking or reasoning needs a dense representation like text or maybe some latent but pure speech is a fuck all dream.

i saw dessa a few days ago. it seems good and sounds good but god knows how much data they trained it on and even it needs text, or some good engineering like ASR over its thinking speech, to actually interact with the real world or be even remotely useful. train on all the call centre data u want then ask it a basic question.

this also reminds me of a question from @juliarturc something like: if you converted every text available into speech and trained on that, what would happen? idk it might be smart or it might not. it depends on a lot of things like whether the speech encoder you use preserves everything needed for speech understanding. codecs are good enough these days but i believe they are still a bottleneck.

even if you solve that,context window is a problem and so is context degradation in speech compared to text LLMs current speech models have the memory of a goldfish.

there are just too many bottlenecks but at the rate we are going we might get there next year only if a lot of people come out in the open and open source the data they are sitting on and stop giving fake demos by training moshi on their privately acquired data.

[❥┊ Akeideu ₊˚. @akeideu]

Somos la primera nación open source ❤️‍🩹

[arXivExpress Speech Generation @arXivExpress]

An LLM Is All You Need: Data-Efficient Adaptation to Full-Duplex, Word-Level Spoken Dialogue

凍結したマルチモーダルLLMを軽量アダプタと離散トークンで適応し、少ないデータで語単位の全二重音声対話を実現した研究。

https://t.co/HzWPfuh8ry

[pulsatingGenius @Neuralwarlock]

the main problem with current full duplex models is that almost everyone is iterating on the same moshi architecture which imo has some real bottlenecks.

in moshi every text token is locked 1:1 to an audio frame that gets spoken right away (apart from padding and special tokens). because of this rigid lockstep the native architecture cant generate a burst of text without advancing audio time. if the model generates 20 text tokens time has to move forward 20 frames so exactly 1.6 seconds of audio gets synthesized alongside. you cant pause the audio clock to let the LLM write out a block of text.

within base moshi you could in theory train it to output reasoning text or JSON tool calls while forcing the audio stream to output silence tokens. but since the clock cant be decoupled that "silent thinking" still runs at 12.5 tokens per second. a 50 token thought chain or tool call means the user sits through 4 solid seconds of awkward dead air.

the other option is synthetic data where it keeps talking while producing the JSON or the thinking. but there is no data in the wild for that so you generate a lot of it and now your data is the bottleneck. the voice will only ever be as natural as the most natural TTS you can use it will always carry the quirks of the harness you built to create it and if whoever is making that data cant think through the scenarios of dyadic conversations well, you are cooked.

so people came up with delegation system which is a pain in itself. you train the frontend model to be smart enough to hold a decent conversation and to hand off tool calling or thinking to a much larger, more capable model. this is engineering and its the only thing that has made the 2 or 3 full duplex models we have like openai gpt live possible at all.

thinking cant be done in speech. the signal to noise ratio is low and there isnt enough speech data to give the model a good enough world model in its weights. how much data is there of people doing maths out loud? even we humans dont think in speech. yes there is "thinking out loud" but speech is not the medium humans think in. thinking or reasoning needs a dense representation like text or maybe some latent but pure speech is a fuck all dream.

i saw dessa a few days ago. it seems good and sounds good but god knows how much data they trained it on and even it needs text, or some good engineering like ASR over its thinking speech, to actually interact with the real world or be even remotely useful. train on all the call centre data u want then ask it a basic question.

this also reminds me of a question from @juliarturc something like: if you converted every text available into speech and trained on that, what would happen? idk it might be smart or it might not. it depends on a lot of things like whether the speech encoder you use preserves everything needed for speech understanding. codecs are good enough these days but i believe they are still a bottleneck.

even if you solve that,context window is a problem and so is context degradation in speech compared to text LLMs current speech models have the memory of a goldfish.

there are just too many bottlenecks but at the rate we are going we might get there next year only if a lot of people come out in the open and open source the data they are sitting on and stop giving fake demos by training moshi on their privately acquired data.

[立川高校天文気象部 @tamc_net]

部員が撮影したアンドロメダ銀河です！
10時間露光することで都内の光に負けず綺麗に映すことができました！！

[メディア画像欄を見てみないか？(よかじぇ) @AMSM_4kz]

留守番中のチェンとペリカの漫画
#エンドフィールド #SanctuaryofInkCC

[PIC @PIC_jpn]

スネージナヤになってから世界任務に救いが無いのですが

[PIC @PIC_jpn]

助けて
この子好きになっちゃう

[ゆと🌸 @yuto___]

だいたいいつも最大ダメージ出してる

[PIC @PIC_jpn]

論理削除がはびこる理由の大半は判断力と知識の不足だと思う。
何かあったらどうするんだで思考停止してるんだよね。
誤削除の対応が必要なら、復旧用DBたてて、バックアップからPITRすればいいのよ。

[IT技術屋のボヤキ @IT_BOYAKI]

「なぜ日本の業務システムは論理削除だらけなのか」という動画、コメント欄が120件を超えて設計思想のぶつかり合いになっているのでまとめます。

いいねが一番多かったのは、消せない業界からの証言でした。

「自治体の住民記録や戸籍、税などのシステムは、統計や過去の履歴に遡っての調査、監査などがあるから、論理削除が基本なのだ。販売管理や在庫管理みたいなシステムを扱ってたところから転職したときは面食らったけど、餅は餅屋、特性に応じての設計が大事だと思った次第なのだ」

次に多かったのが、消される側の感覚。

「ユーザー、特に非エンジニアの人からすれば、退会（削除）したつもりのサービスから自分の個人情報が流出したとなったら『は？』ってなりそうだよなぁ」

「情報流出が頻発してる昨今だと、個人情報の削除を求めたのに個人情報が漏れた、って一段上のやばい事態になる気がする」

設計者にとってのフラグ1本が、ユーザーにとっては「消したはずのもの」です。

現場で見た実装の話では、これが強烈でした。

「今まで見た中でいちばんびっくりしたのは、削除フラグすら存在しない論理削除かなあ。論理削除実行時にidの数値符号を反転させて画面から消す」

フラグ方式そのものへの不満も根強いです。

「論理にするならゴミ箱用の履歴テーブルに入れれば良い派。フラグ管理って嫌いなんだよなぁ。このフラグ一つでテーブルの性能も格段に落ちるし」

「論理削除や履歴が同じテーブルにあると、最新データを取得することに対してのルールが必要になるが、漏れが多く、必ず問題になる。それに反して、本来想定している誤削除の対応は言うほどない」

「論理削除で3年後にDBアップグレードでいらない予算を溶かすのを何回も見ているなぁ」

一方で、「とりあえず論理削除」を擁護する冷静な意見もありました。

「『とりあえず論理削除』は安全の意図もあるけど、それ以上にコスパの都合もある。問題になるときって大体運用もハッキリ見えてない。そのタイミングで、本題では無い削除の挙動の話を詰めるより、とりあえず論理削除で進める方が早い」

「データベースの寿命が長過ぎて、大昔に作ったものと今言われてるベストプラクティスがズレてることがよくある」

落としどころとして多かったのは、両方使う運用です。

「ISMSの関係で両方使う。例えば、会員が退会したら論理削除。1ヶ月経過して何もなければ論理削除したものを物理削除」

そして、論理削除が増える本当の理由。

「マネージャー『もし物理削除したことで何か起きたら、お前責任取れるんだろうな！』この手のやり取りで、どれほどのバグやトラブルが起きているだろうか…」

最後に、一番短くて一番刺さったやつ。

「ソースコードを論理削除してる所はデータも論理削除している」