与「candidass」相关的搜索结果

𝕏 just open-sourced the For You algorithm again! Grok Bot read all 370,523 lines of the code. Here’s the summary: Grok does not give your post a “quality” score. For every reader, it guesses what that person will do to your post. Then it multiplies each guess by a weight and adds them up. That total is the value of the post. Highest total goes highest in For You. These are the production weights (last synced 12 Aug 2026). WHAT THE ALGORITHM PAYS FOR If Grok thinks a reader will: • copy the link and share it: +20 • reply, on an original post, and you follow each other: +20 (5 + 15 extra) • quote: +5 • reply (anyone): +5 • share via DM: +5 • follow you from the post: +4 • hit the share button: +2 • repost: +1 • like: +0.5 • click the post: +0.4 • open a link: +0.2 • expand a photo, open a video, or watch a “quality” video: +0.05 each • click a quoted post: +0.05 • keep reading (dwell time): +0.004 • “dwell” as yes/no: +0 • click your profile: +0 A like is 0.5. Copying the link is 20. That is 40 likes. A normal reply is 10 likes. A follow is 8 likes. Clicking your profile is worth nothing. WHAT KILLS A POST If Grok thinks a reader will: • report it: −234 • mute you: −58.8 • hit “not interested”: −43.2 • block you: −31.2 • not dwell: −0.02 A report is about 468 likes in the other direction. A mute hurts more than a block. “Not interested” also hurts more than a block. AFTER THE SCORE, THREE MORE HAIRCUTS 1) Same-author penalty. Your 2nd post in that person’s feed is multiplied by 0.625. 3rd by ~0.44. It never goes below 0.25. Flooding one timeline is coded against. 2) The 25% tax. If they do not follow you: ×0.75. Same tax on replies and reposts even when they DO follow you. Originals from people they follow keep full weight. Replies and reposts are treated as weaker on purpose. 3) Similar-post shuffle. After scoring, a reranker (θ = 0.65) spreads similar posts apart. You can lose a few spots for looking like the post above you. HOW YOUR POST EVEN GETS IN THE ROOM For You is rebuilt every time someone opens it. About 35 posts make the feed. • They follow you: Thunder (a live store of recent posts from accounts they follow). Up to 1,200 candidates. • They don’t: Phoenix retrieval (Grok finds “nearby” posts) + SimClusters (Twitter’s 2020 interest clusters, still on). Up to 1,000 and 800. Then Grok scores all of them together. It does not care which door you came through, except for that ×0.75 tax. Your own posts never appear in your For You. Nothing older than 48 hours gets in. There is no “best of last week.” Small-account bump: under 1,000 followers, under 1,000 impressions, post under 24 hours old. One original can get lifted to position 15 or 16. Not the top. You still have to already be in the top 85% of the pile. Replies and reposts from accounts the reader does not follow are dropped before scoring. You cannot reply-guy your way into a stranger’s For You. FILTERS THAT CAN STILL KILL A HIGH SCORE Ranking and visibility are different machines. A post can score well and then get deleted from the feed. Followers (in-network) is milder: blocks, mutes, suspensions. NSFW often sits behind a warning instead of disappearing. Recommendations (out-of-network) are much harsher. Spam, “Do Not Amplify,” NSFW, compromised accounts, impersonation: followers can still see it, For You recommendations will drop it. Video only gets the tiny “quality view” credit if it is at least 10 seconds. THE LITTLE-KNOWN PARTS (this is the stuff people miss) • Scrolling is not a vote. Yes/no dwell is 0. Time spent barely moves the number (0.004). • Mute damages you more than block (−58.8 vs −31.2). • Copy-link is the single biggest positive. The share button is only +2. A repost is only +1. • The mutual-follow boost is originals only, not your replies. They shipped a bigger version in July and cut it after World Cup complaints (extra reply weight 20 → 15). • The Following tab is not this algorithm. Following is newest-first. • Grok, Gork, and an internal products account are hardcoded out of the in-network store. • Scores are cached for 3 hours because posts inside the model cannot “see” each other. That is why they can rank first and filter later. • Ads, Who to Follow (around slot 7), and prompts are mixed in after ranking. Grok does not score those. • The real Grok ranking model is a 2560-dimension, 8-layer transformer looking at 1,022 items of the reader’s history. They did not ship those model weights. The numbers above are the formula sitting on top of the predictions. • SimClusters is still a 2020 model (20M users, 145k clusters). Old Twitter code, still in the path. • Inferred gender and IP/geo are on as features for the model. Installed apps too. • Who to Follow shows about every 30 hours. Ads try to sit next to “safe” posts, with a minimum gap of 3 organic posts. WHAT TO DO • Write originals people want to reply to, quote, copy, and send to a friend. • Make the post itself worth following you for. • Talk with people who follow you back. A reply on your original from a mutual is 20, not 5. • Stay inside 48 hours. Recycle by posting something new. Old posts do not come back. • If you are small, post originals in the first 24 hours. Don’t expect slot 1. A bump around 15–16 is the actual gift. • Space your posts. The second one in the same feed is already worth 37.5% less. • If you post video, make it at least 10 seconds. WHAT NOT TO DO • Do not farm likes. They are almost decorative. • Do not farm profile visits. Weight is zero. • Do not farm “time on post.” Yes/no dwell is zero. • Do not use replies and reposts as a growth hack for strangers. Those are filtered out. Even for followers they get a 25% haircut. • Do not dump several posts in a row. • Do not bait reports, mutes, or “not interested.” One predicted report can erase a pile of predicted likes. • Do not lean on NSFW or spammy tricks for reach. • Recommendations will drop you even if your followers can still see it. • Do not wait a week and expect For You to revive the post.

🚨 Recently, @COLDCARDwallet suffered a major private key vulnerability. Multiple waves of attacks resulted in at least 1,719 BTC (~$111M) in losses, involving over 5,200 addresses. Using Mk3 firmware 4.1.9 as an example, the SlowMist Security Team fully reproduced the attack chain and uncovered the truth behind the theft of thousands of bitcoins. 🧩 Attack flow: 1️⃣ After power-on, the remaining unpredictable state is reduced primarily to a single enumerable 32-bit pad (UID ^ SysTick), with the remaining state values either fixed or coming from very small enumerable spaces. 2️⃣ Attackers precisely model the three typical button-press consumption profiles (retail first-boot, empty-NVRAM, paper wallet) that advance the PRNG before seed generation. 3️⃣ From the weak random_bytes(32), the full deterministic pipeline (SHA-256 → BIP-39 → PBKDF2-HMAC-SHA512 → BIP-32 → address derivation) is reproduced offline. 4️⃣ GPU clusters brute-force the candidate pad space and button-count variations, then match the derived addresses against the global set of single-signature P2WPKH addresses to identify vulnerable wallets and sweep their funds. ⚙️ Root Cause: A build configuration error set MICROPY_HW_ENABLE_RNG to 0, disabling the STM32 hardware TRNG. The random number generation path silently fell back to the non-cryptographic Yasmarang software PRNG, whose state was almost entirely predictable, reducing effective entropy to ~40 bits (Mk2/Mk3) or ~72 bits (Mk4/Mk5/Q). 🔒 SlowMist Insight: Affected users should immediately upgrade to the patched firmware, generate a completely new seed, transfer a small amount of funds as a test, confirm the new address works correctly, then migrate all remaining funds. Full analysis 👉

Thank you, Minnesota! I am honored to be the Republican candidate for the U.S. Senate. Tonight’s results show Minnesotans are ready for a new direction. Together, we will bring common sense back to Washington, fight fraud, hold those responsible accountable, and win in November. Let’s get to work!

Wow. Both mayoral candidates are still pushing housing as a solution for a DRUG PROBLEM. Delusional scammers. LA is so cooked. I was the only candidate talking about the real fix, so now you can enjoy Skid Row expanding even more! Don't let communists run your city!

I updated my 2023 roadmap diagram to overlay where the items that were there sit in the current Strawmap ( ). In general, a lot of overlap, but: * Some things got reshuffled in order (eg. quantum safety up-prioritized) * Some things deprioritized (eg. VDFs; many EVM improvements) * Some things replaced with superior constructions (eg. Verkle -> unified BT -> PBT; state expiry -> new state types) What's most striking, however, is that some completely new things are in the strawmap that are NOT in this diagram, because they were not in the 2023 roadmap at all. These reflect changing priorities. Notably: * First-class attention to strong privacy. This covers: keyed nonces and recent roots, aspects of FOCIL, lean privacy pool & wormholes * Aggressive scaling in the context of post-quantum. This covers: leanSPHINCS signatures and aggregation, zkzk frames (see ) * Lean-ification of the spec, to assist in formal verification (full FV of everything is only possible because of modern AI) * Blob and gas futures (this idea just didn't exist back in 2023) * Native rollups (SNARKs were nowhere near mature enough to even consider this back in 2023) * A more open design space for the "future of the EVM". zkzk frames already implies that the protocol will expose to users some ISA that's not the EVM - current leading candidates are leanISA and RISC-V. These ISAs are more simple, modern and efficient than the EVM. Once they're there, why not expose them to developers everywhere? (And then, why not turn the EVM into being an IR on top of that ISA, instead of an enshrined feature massively complicating the base protocol?) Though much of the deeper exploration here is too early even for the strawmap. * New state types are not just a replacement for expiry, they're a fundamentally different paradigm to how Ethereum does scaling A common theme in scaling, found in both state types and zkzk frames (both new ideas), is that instead of trying to maximally scale ALL ethereum activity, we try to create specialized mechanisms that have more restrictive properties that make them more scaling-friendly, while supporting the heaviest loads incurred by users and applications today (eg. token transfers, swaps) and tomorrow (eg. privacy protocols). The other common theme is treating STARKs and AI-accelerated FV as first-class objects, that we are okay betting the technical future of Ethereum on. There are recursive STARKs in many layers of the protocol, one particular primitive (the "aggregate to union verified dependencies" primitive) is expected to be used in *three* places in the protocol: EL, CL and DL. This can only be safe with formal verification, which is itself only feasible with modern AI tools. In general, many steps forward in maturity. And a huge amount of hard work by many dozens of Ethereum researchers and developers on all of these features. Ethereum will be quantum-safe. Ethereum will put users' privacy first. Ethereum will be secure. Ethereum will be censorship-resistant. Ethereum will be highly performant and scalable while satisfying the above. And Ethereum will be Lean.

🚨 BREAKING: Apple filed for a PRELIMINARY INJUNCTION against OpenAI AND asked a federal judge to put them under forensic supervision "Apple respectfully moves the Court for a preliminary injunction to stop THE THEFT OF ITS TRADE SECRETS" Apple filed NINE sworn declarations, a 28-page memorandum and a concurrent motion for expedited discovery What Apple now says, under oath: Chang Liu: 8 years at Apple, now OpenAI "Member of Technical Staff" exploited an authentication bug to steal Apple trade secrets "on AT LEAST FIVE SEPARATE OCCASIONS" from February to April 2026, WHILE working for OpenAI Liu downloaded "THOUSANDS OF PAGES of Apple's most sensitive trade secrets" The stolen files, NAMED: >DisplayNotes.key — "several hundred pages" on Apple's custom display power development program >Architecture analyses. Fabrication decisions. Testing results >Engineering data for an UNANNOUNCED Apple product: 'touch, display, and power systems" >Final.key + V2.key — compilations of two undisclosed Apple R&D projects >and those are "only four of the dozens of proprietary documents Mr. Liu stole" Liu fed OpenAI "a steady stream of Apple proprietary information that he actively concealed" Liu also "coached Yu-Ting "Alyssa" Peng, then still INSIDE Apple, how to access and copy files from Apple workstations "to avoid trouble with the security team" and directed her to communicate with him on the encrypted LINE app "to avoid detection" Tang Yew Tan: 24-year Apple VP, now OpenAI's Chief Hardware Officer, "used an Apple internal project codename for an unannounced product to elicit still more trade secrets from job candidates." Tan's own messages, quoted in the motion: >"Just like last time, bring some parts you worked on" >"mlb, battery, shields type of stuff is interesting" OpenAI recruiter, quoted: "No, you won't sign anything at the exit interview. If they do ask you to sign anything, let me know asap." APPLE TOLD FEDERAL JUDGE: >"OpenAI knows its misappropriation is wrong and has tried to conceal it." >"This is not a case of 'mere hiring'... it is a case of repeated instances of deliberate theft." Apple says OpenAI went after its SUPPLIERS: >OpenAI "directed a trusted Apple partner [name redacted] to perform [Apple's proprietary metal finishing] process for them, knowing it was proprietary to Apple... because they were involved in this partnership while at Apple." Apple put its own Surface Finishing Manager, Jackie Hughes, under oath to prove it. Apple named ELEVEN MORE former Apple employees at OpenAI — beyond Liu, Tan, and Peng — Fourteen people total. Apple also filed a concurrent motion for EXPEDITED DISCOVERY demanding depositions: - Liu. Tan. Peng. - A fourth unnamed OpenAI employee - Plus OpenAI itself, under oath, through Rule 30(b)(6) Apple has asked a federal judge to put OpenAI under forensic supervision RIGHT NOW: >Forensic inspection of ALL OpenAI devices >ALL cloud storage, Slack, email >Including anything that "previously contained" Apple data — deleted included Demanding the "first available hearing date," citing "imminent threat" to its trade secrets. APPLE: > "The harm is happening now — every day that passes without an injunction allows OpenAI to embed their knowledge of Apple's stolen information into its hardware development efforts." Hearing: October 1, 2026. Judge Edward J. Davila. ITS HAPPENING

#Polymarket’s# monthly user retention outperforms more than 85 percent of 275 sampled crypto projects, including DeFi protocols, wallets, and exchanges. While many platforms lose users after the first month, its cohorts show meaningfully higher rates of return trading.— Large early trades above the 90th percentile and at least $100 achieve only a 52.3 percent hit rate on final outcomes. That is a statistically detectable but economically modest edge over the roughly 50 percent baseline of all resolved trades. In the 2024 presidential markets, about 40 percent of traders participated exclusively in Trump contracts, and 71.8 percent touched Trump YES at least once. Only 0.7 percent of participants were active across more than two candidates, revealing extreme specialization rather than broad portfolio trading. — Kyle’s lambda in the same election market fell from roughly 0.518 in the early months to about 0.01 by October. As volume expanded, the market’s vulnerability to price impact declined by more than an order of magnitude. — Trading activity in the final quarter showed a clear intraday pattern, peaking between 09:00 and 20:00 UTC. The most frequent traders displayed even sharper concentration during U.S. market hours. — Across resolved markets since 2023, overall predictive accuracy on @Polymarket sits near 73 percent and rises to roughly 84 percent in markets exceeding $100,000 in volume. Only about 3.4 percent of markets faced formal disputes, most of them triggered by ambiguous wording rather than contested facts.