正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

By now, everyone knows that single-vector embedding models are hugely limiting for modern workflows. But they contain than you think: you can extract sparse Latent Terms from them. And it turns out that BM25 is all you need to turn this vocabulary into a strong retriever.

Mixedbread

3,415 subscribers

40,023 次观看 • 1 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Everyone knows this Ren and Stimpy animation for the "Don't reference modern shit, you idiot" scene, but I think this joke that comes before it is pretty underrated. I quote this to myself a lot lol.

Everyone knows this Ren and Stimpy animation for the "Don't reference modern shit, you idiot" scene, but I think this joke that comes before it is pretty underrated. I quote this to myself a lot lol.

Kat On A Mat 🆖

261,366 次观看 • 10 个月前

They think it’s a joke that you’re paying for their lifestyle, they think you owe it to them. THIS is the attitude you see all over the internet from people who are on food stamps. It sure looks like they’re just as able bodied as anyone. If you can dance you can work!

They think it’s a joke that you’re paying for their lifestyle, they think you owe it to them. THIS is the attitude you see all over the internet from people who are on food stamps. It sure looks like they’re just as able bodied as anyone. If you can dance you can work!

One Bad Dude

822,378 次观看 • 8 个月前

This finishing concept is great when you’re not in a blow by situation. Skip into them so that they are forced to absorb your contact. When you land and you want to transition your way to your back foot so that you can feed without being out of control

This finishing concept is great when you’re not in a blow by situation. Skip into them so that they are forced to absorb your contact. When you land and you want to transition your way to your back foot so that you can feed without being out of control

Reid Ouse

99,092 次观看 • 1 年前

Big moment for Postgres! Search has always been Postgres' weak spot, and everyone just accepted it. If you needed a real relevance-ranked keyword search, the default answer was to spin up Elasticsearch or add Algolia and deal with the data sync headaches forever. The problem isn't that Postgres can't do text search. It can. But the built-in `ts_rank` function uses a basic term frequency algorithm that doesn't come close to what modern search engines deliver. So teams end up: - Running a separate Elasticsearch cluster just for search - Building sync pipelines that inevitably drift out of consistency - Paying for managed search services that charge per query - Accepting mediocre search relevance because "good enough" ships faster But this is actually a solvable problem. You can realistically bring industry-standard search ranking directly into Postgres, which eliminates the need for external infra entirely. This exact solution is now available with the newly open-sourced pg_textsearch by Tiger Data - Creators of TimescaleDB, a Postgres extension that brings true BM25 relevance ranking into the database. BM25 is the algorithm behind Elasticsearch, Lucene, and most modern search engines. Now it runs natively in Postgres. Here's what pg_textsearch enables: - True BM25 ranking with configurable parameters (the same algorithm powering production search systems) - Simple SQL syntax: `ORDER BY content 'search terms'` - Works with Postgres text search configurations for multiple languages - Pairs naturally with pgvector for hybrid keyword + semantic search That last point matters a lot for RAG apps. The video below shows this in action, and I worked with the team to put this together. You can now do hybrid retrieval (combining keyword matching with vector similarity) in a single database, without stitching together multiple systems. The syntax is clean enough that you can add relevance-ranked search to existing queries in minutes. pg_textsearch is fully open-source under the PostgreSQL license. You can find a link to their GitHub repo in the next tweet.

Big moment for Postgres! Search has always been Postgres' weak spot, and everyone just accepted it. If you needed a real relevance-ranked keyword search, the default answer was to spin up Elasticsearch or add Algolia and deal with the data sync headaches forever. The problem isn't that Postgres can't do text search. It can. But the built-in `ts_rank` function uses a basic term frequency algorithm that doesn't come close to what modern search engines deliver. So teams end up: - Running a separate Elasticsearch cluster just for search - Building sync pipelines that inevitably drift out of consistency - Paying for managed search services that charge per query - Accepting mediocre search relevance because "good enough" ships faster But this is actually a solvable problem. You can realistically bring industry-standard search ranking directly into Postgres, which eliminates the need for external infra entirely. This exact solution is now available with the newly open-sourced pg_textsearch by Tiger Data - Creators of TimescaleDB, a Postgres extension that brings true BM25 relevance ranking into the database. BM25 is the algorithm behind Elasticsearch, Lucene, and most modern search engines. Now it runs natively in Postgres. Here's what pg_textsearch enables: - True BM25 ranking with configurable parameters (the same algorithm powering production search systems) - Simple SQL syntax: `ORDER BY content 'search terms'` - Works with Postgres text search configurations for multiple languages - Pairs naturally with pgvector for hybrid keyword + semantic search That last point matters a lot for RAG apps. The video below shows this in action, and I worked with the team to put this together. You can now do hybrid retrieval (combining keyword matching with vector similarity) in a single database, without stitching together multiple systems. The syntax is clean enough that you can add relevance-ranked search to existing queries in minutes. pg_textsearch is fully open-source under the PostgreSQL license. You can find a link to their GitHub repo in the next tweet.

Akshay 🚀

215,344 次观看 • 5 个月前

You think that you are a strong man 🥱 But I know how to manipulate and take everything! My feet drives you crazy. Active now for online sessions 💅🔔

You think that you are a strong man 🥱 But I know how to manipulate and take everything! My feet drives you crazy. Active now for online sessions 💅🔔

Mistress Raika Macabra•

10,720 次观看 • 2 年前

If you encore a Prankster Pokemon into a the previous move and that move was an attacking move but this turn they went for a Status move, then it will get the Prankster priority. But now that also means that it won’t hit dark-type Pokemon that turn.

If you encore a Prankster Pokemon into a the previous move and that move was an attacking move but this turn they went for a Status move, then it will get the Prankster priority. But now that also means that it won’t hit dark-type Pokemon that turn.

Gio™

111,793 次观看 • 2 个月前

multi-turn RL and the "tito" problem keeps coming up. we've been working on it for a while, and the takeaway is that it's much easier than people are making it. it takes 1 implementation rule, and 1 chat-template property that all models already comply with. **that's all you need to do it right**

multi-turn RL and the "tito" problem keeps coming up. we've been working on it for a while, and the takeaway is that it's much easier than people are making it. it takes 1 implementation rule, and 1 chat-template property that all models already comply with. that's all you need to do it right

Quentin Gallouédec

34,780 次观看 • 1 个月前

🔜 Duke Neurosurgery Thank you to everyone that stood by me during my wildest dreams. This is only the beginning, and to all that are watching: if I can, then you can too. #Match2024 #Neurosurgery

🔜 Duke Neurosurgery Thank you to everyone that stood by me during my wildest dreams. This is only the beginning, and to all that are watching: if I can, then you can too. #Match2024 #Neurosurgery

Julia Duvall

19,415 次观看 • 2 年前

“Paano ba ‘to from now on?” I wasn’t expecting they would drop that. I can’t even type it out because I don’t want it to be real, but it is what it is, and I fully respect that. It was a hard decision, for sure, and yes, it is scary if you look at it from here on, but I am all for their growth. It brings me comfort that that decision was made to grow more individually as an artist and as a person. I’m just so proud that they were brave enough to take on that, knowing all the uncertainties that are laced into that decision. As a fan, it is still hard to process all of that for now, but I know that no matter what happens, they will always have each other and nobody can take away what they have and what they have built throughout the years. Nothing will ever change. I’m also grateful that it came straight from them, not from anyone else. It just shows how they love, respect, and care for us, their bubblies. So yeah, donbelle’s last project for now, but please know that there will always be a house you can come home to anytime. ‘Wag masyadong magpa-miss ah, Belle Mariano Donny. Mahal na mahal ko kayo 🤍

“Paano ba ‘to from now on?” I wasn’t expecting they would drop that. I can’t even type it out because I don’t want it to be real, but it is what it is, and I fully respect that. It was a hard decision, for sure, and yes, it is scary if you look at it from here on, but I am all for their growth. It brings me comfort that that decision was made to grow more individually as an artist and as a person. I’m just so proud that they were brave enough to take on that, knowing all the uncertainties that are laced into that decision. As a fan, it is still hard to process all of that for now, but I know that no matter what happens, they will always have each other and nobody can take away what they have and what they have built throughout the years. Nothing will ever change. I’m also grateful that it came straight from them, not from anyone else. It just shows how they love, respect, and care for us, their bubblies. So yeah, donbelle’s last project for now, but please know that there will always be a house you can come home to anytime. ‘Wag masyadong magpa-miss ah, Belle Mariano Donny. Mahal na mahal ko kayo 🤍

nato&tinky🧚

263,480 次观看 • 2 个月前

I’m a fan of Aunty Toyin Abraham but this her unnecessary kneeling down and greeting is becoming too much and looking like an eye service now. It’s ok to respect elders and cultures but those women she keep kneeling for are not so older than her. Aunty Toyin, you need to start packaging yourself now. That’s why they keep disrespecting you. They are not more than you, they are not bigger than you. Stop belittling yourself for them. We Nigerians have forgiven you for supporting City Boy in 2023. It’s time you forgive yourself too and carry yourself with pride and grace cos you work hard for it and you deserve the respect that comes with it. I don’t mean any disrespect and I love you. I hope this gets to you TOYIN ABRAHAM AJEYEMI

I’m a fan of Aunty Toyin Abraham but this her unnecessary kneeling down and greeting is becoming too much and looking like an eye service now. It’s ok to respect elders and cultures but those women she keep kneeling for are not so older than her. Aunty Toyin, you need to start packaging yourself now. That’s why they keep disrespecting you. They are not more than you, they are not bigger than you. Stop belittling yourself for them. We Nigerians have forgiven you for supporting City Boy in 2023. It’s time you forgive yourself too and carry yourself with pride and grace cos you work hard for it and you deserve the respect that comes with it. I don’t mean any disrespect and I love you. I hope this gets to you TOYIN ABRAHAM AJEYEMI

Lady Victoria

227,206 次观看 • 3 个月前

✨ Made a new mini feature on Photo AI: [ Grab from 3d model ] So the problem is we're at that stage in time (typical for AI) where image-to-3d models are not good enough but are fun to play with, but we know they'll be good enough in 1-2 years With [ Make 3d model ] you already can turn any Photo AI pic into a 3d model but it still looks hyper clunky and deformed, but it works! One cool idea I had to make that more useful and made now: Let people make a 3d model then change the view of the it with the 3d viewer, then press [ o ] and it grabs a frame of the 3d That image you can then [ Remix ] (img2img), and it becomes a real photo again and that in turn you can then turn into a video again with [ Make video ] So that essentially gives you a fully freeform camera position control to take photos with One thing I need to fix is the background/skybox, I kinda need to take the original photo and remove the person and just get the background for the 3d model viewer, in this case it should be white, but it's a start!

✨ Made a new mini feature on Photo AI: [ Grab from 3d model ] So the problem is we're at that stage in time (typical for AI) where image-to-3d models are not good enough but are fun to play with, but we know they'll be good enough in 1-2 years With [ Make 3d model ] you already can turn any Photo AI pic into a 3d model but it still looks hyper clunky and deformed, but it works! One cool idea I had to make that more useful and made now: Let people make a 3d model then change the view of the it with the 3d viewer, then press [ o ] and it grabs a frame of the 3d That image you can then [ Remix ] (img2img), and it becomes a real photo again and that in turn you can then turn into a video again with [ Make video ] So that essentially gives you a fully freeform camera position control to take photos with One thing I need to fix is the background/skybox, I kinda need to take the original photo and remove the person and just get the background for the 3d model viewer, in this case it should be white, but it's a start!

@levelsio

119,210 次观看 • 1 年前

Next time you’re looking for a book to read, a show to watch, or an item to shop for, all it takes is a single prompt. By activating Personal Intelligence, Gemini can connect to your Google apps and give you suggestions that are customized for you.

Next time you’re looking for a book to read, a show to watch, or an item to shop for, all it takes is a single prompt. By activating Personal Intelligence, Gemini can connect to your Google apps and give you suggestions that are customized for you.

Google Gemini

92,115 次观看 • 6 个月前

God has made you stronger than anything that will come against you. There are things that you would think would break you: the sickness, the addiction, the trouble, the loss. It feels bigger, stronger, tougher, but can I encourage you: you are unbreakable. Listen to this message from Joel! "Still I Rise":

God has made you stronger than anything that will come against you. There are things that you would think would break you: the sickness, the addiction, the trouble, the loss. It feels bigger, stronger, tougher, but can I encourage you: you are unbreakable. Listen to this message from Joel! "Still I Rise":

Joel Osteen

22,530 次观看 • 3 个月前

To all those who are bone tired, struggling, or quietly wondering if you are enough: You are probably giving yourself far less credit than you deserve. You may not see it, but your presence matters. Your efforts matter. The kindness you show, the burdens you carry, the way you keep showing up and doing your best each day has an impact on more people than you know. And to everyone else, never underestimate the power of reminding someone that they are seen, appreciated, and loved. A few sincere words can lighten a burden, restore hope, and change the course of a day. We all need that reminder sometimes. op: Ring

To all those who are bone tired, struggling, or quietly wondering if you are enough: You are probably giving yourself far less credit than you deserve. You may not see it, but your presence matters. Your efforts matter. The kindness you show, the burdens you carry, the way you keep showing up and doing your best each day has an impact on more people than you know. And to everyone else, never underestimate the power of reminding someone that they are seen, appreciated, and loved. A few sincere words can lighten a burden, restore hope, and change the course of a day. We all need that reminder sometimes. op: Ring

Cory Booker

20,611 次观看 • 1 个月前

Anyway. Look at this Barrett i own for marketing (that was literally my reason for buying it) So. Look at it. You want to buy a fuk-bin at 3pm pacific now Or you don’t because you are a broke nigga who knows you don’t need another random order but look at that Barrett man. How cool.

Anyway. Look at this Barrett i own for marketing (that was literally my reason for buying it) So. Look at it. You want to buy a fuk-bin at 3pm pacific now Or you don’t because you are a broke nigga who knows you don’t need another random order but look at that Barrett man. How cool.

PNWGUERRILLA

27,123 次观看 • 3 个月前

the people running shopify stores are being robbed. they tell you that you need apps for everything. they tell you that you need a monthly bill for every little thing. it is a scam. my stores do not have apps. my stores use sections. it is very fast. it is very clean. you buy it one time and you own it forever. you can use it on all your stores. the app guys hate me for this. they want your money every month. i want you to win. i am giving away one of my top sections today. comment "win" below. i will send it to you. (you must follow me so i can send the link)

the people running shopify stores are being robbed. they tell you that you need apps for everything. they tell you that you need a monthly bill for every little thing. it is a scam. my stores do not have apps. my stores use sections. it is very fast. it is very clean. you buy it one time and you own it forever. you can use it on all your stores. the app guys hate me for this. they want your money every month. i want you to win. i am giving away one of my top sections today. comment "win" below. i will send it to you. (you must follow me so i can send the link)

George

19,423 次观看 • 5 个月前

Dear Tesla AI, I am on I-95 South near Washington DC. I just disengaged my Tesla because I thought it was confusing the shoulder for a lane. I logged the disengagement as critical. As you can see in the video below, it turns out that once again, I’ve mistakenly assumed my meat brain knows better than your AI model. I wanted to let you know that this disengagement can be removed from the statistics, as it once again turns out that I am the fool. Sorry about that. Thank you for reading.

Dear Tesla AI, I am on I-95 South near Washington DC. I just disengaged my Tesla because I thought it was confusing the shoulder for a lane. I logged the disengagement as critical. As you can see in the video below, it turns out that once again, I’ve mistakenly assumed my meat brain knows better than your AI model. I wanted to let you know that this disengagement can be removed from the statistics, as it once again turns out that I am the fool. Sorry about that. Thank you for reading.

Mike P

283,811 次观看 • 2 天前