tshanmu
Re:Three Days in August: What a DDoS Attack Exposed in Our Network
on Monday September 28, 2026 @07:29AM (#49876407)
"We found three concrete gaps during this incident, and we would rather be upfront about them than gloss over them." claudism?
Re:Trees use a "muscle", tension wood, to correct their posture
on Wednesday September 09, 2026 @04:56AM (#49623432)
Re:Splash-free urinals (2025)
on Monday September 07, 2026 @12:13PM (#49599942)
a surprising paragraph in the middle of the article "The “urine-no,” a hostile antiurination surface"
Re:Telegraph censured for story of fictional family's struggle to pay ...
on Tuesday March 03, 2026 @03:10PM (#47238222)
In the UK, the telegraph newspaper censured for story of fictional top-earner family’s struggle to pay school fees
Re:I wanted a camera that doesn't exist – so I built it
on Tuesday January 06, 2026 @11:51AM (#46514784)
Wow! This is the kind of news that we want on hacker news in 2026. Amazing hacker ethos.
The article could have been better with sharing some photos taken with the new camera.
Re:Beaver: An Efficient Deterministic LLM Verifier
on Tuesday December 16, 2025 @10:33AM (#46289800)
As large language models (LLMs) transition from research prototypes to production systems, practitioners often need reliable methods to verify that model outputs satisfy required constraints. While sampling-based estimates provide an intuition of model behavior, they offer no sound guarantees. We present BEAVER, the first practical framework for computing deterministic, sound probability bounds on LLM constraint satisfaction. Given any prefix-closed semantic constraint, BEAVER systematically explores the generation space using novel token trie and frontier data structures, maintaining provably sound bounds at every iteration. We formalize the verification problem, prove soundness of our approach, and evaluate BEAVER on correctness verification, privacy verification and secure code generation tasks across multiple state of the art LLMs. BEAVER achieves 6 to 8 times tighter probability bounds and identifies 3 to 4 times more high risk instances compared to baseline methods under identical computational budgets, enabling precise characterization and risk assessment that loose bounds or empirical evaluation cannot provide.
Beaver: An Efficient Deterministic LLM Verifier
Re:COBOL to Kotlin via Formal Models (IR and Alloy and Golden Master)
on Thursday November 13, 2025 @01:27PM (#45918556)
how are you creating the IR?
Re:I built ChatGPT with Minecraft redstone [video]
on Thursday October 02, 2025 @07:33AM (#45448335)
this reminds me of permutation city and the question of what is time...
Re:US Airlines Push to Strip Away Travelers' Rights by Rolling Back Ke...
on Wednesday September 24, 2025 @09:58AM (#45360508)
you would know before consuming how much you would end up paying. not so in the land of the free.
Re:Viking-Age hoard reveals trade between England and the Islamic World
on Monday August 18, 2025 @07:51AM (#44939610)
this sounds about right - the same yardstick was used by the British to define Hindus in India - people who are neither christian nor muslim.
Re:Literate programming tool for any language
on Friday June 20, 2025 @11:03AM (#44328385)
wondering how would literate programming be with the actual code being written by LLM? I have been searching for any tools that allow for such a setup... may be my next weekend project in a long list of projects
Re:Millions in west don't know they have aggressive fatty liver diseas...
on Thursday June 05, 2025 @05:08PM (#44195799)
I can concur - that shift in mindset to "i'm hungry, oh well" is crucial for your body I feel.
Re:The Barbican
on Tuesday May 13, 2025 @02:42AM (#43970146)
nice analogy comparing it to BBC Radio 3- if you/someone knows which neighbourhood would be like BBC Radio 4? I find R3 too high brow for me - Radio 4 seems more accessible :)
Re:We're bringing Pebble back
on Wednesday January 29, 2025 @08:33AM (#42864708)
+1 pleasantly surprised by my garmin forerunner for exactly the same with hrv stats as well
Re:Coinbase awarded a $500k bug bounty
on Thursday August 01, 2024 @05:55AM (#41127543)
certik appears to be a web3 smart contract audit company
Re:Hidden Alpha
on Thursday August 01, 2024 @05:10AM (#41127333)
This paper documents the central role of hidden connections between fund managers and firm officers in financial markets, drawing on an extensive dataset of over 100 thousand manually identified Facebook profiles and their 35 million Facebook friends. Our findings reveal that the hidden connections between these individuals are associated with the largest and most significant abnormal returns accruing to fund managers, averaging 135 basis points per month (over 16% alpha per year, t-stat = 3.54) across the universe of mutual funds and public firms. In stark contrast, trades involving publicly visible connections generate no significant abnormal returns on average.
Hidden Alpha
Re:Code Finder – The ultimate search engine for GitHub repositories
on Wednesday June 12, 2024 @12:24PM (#40659874)
Why would I use this instead of github code search, which appears to be quite comprehensive?
Re:One Knuckle Rice Rule
on Tuesday April 16, 2024 @03:51AM (#40049404)
The traditional Indian style of cooking involves cooking till about 80% on heat with however much you want and then drain the water (like pasta). I remember reading somewhere[1] that it reduces the amount of the arsenic leftover in rice by draining the water.
[1] https://scitechdaily.com/scientists-find-new-way-of-cooking-...
Edit: added source
Re:Lo Manh Hung, the Story of the Youngest Photo Journalist in South V...
on Thursday April 11, 2024 @02:26PM (#40005239)
same - I tried unsuccessfully to find his eventual story.
Re:We've saved 5000% in cloud costs by writing our own database
on Sunday April 07, 2024 @02:54AM (#39958762)
"Of course, that’s an unfair comparison, after all, Postgres is a general purpose database with an expressive query language and what we’ve built is just a cursor streaming a binary file feed with a very limited set of functionality - but then again, it’s the exact functionality we need and we didn’t lose any features."
Re:The Race to 2nm: RISC-V Chips in Japan – By Dr. Ian Cutress
on Thursday March 07, 2024 @04:23PM (#39634741)
thanks!
Re:The Race to 2nm: RISC-V Chips in Japan – By Dr. Ian Cutress
on Thursday March 07, 2024 @01:34PM (#39632802)
while it all sounds buzzwordy, as an chip industry outsider, none of it seems to make much sense.. is there an ELI5 version of this?
Re:Brits make Amazon, Meta stop using third-party data to undercut rivals
on Sunday November 05, 2023 @01:25PM (#38153969)
sad that el reg is also doing this:
"In Amazon's case, the e-commerce giant used vendors' sales figures to decide which items it should sell, and how much to price products to get an edge over everyone else. The internet behemoth also promoted its own products with its Buy Box feature and it further cut into retailers' margins by charging extra costs if they wanted to use Amazon's Prime delivery services, the CMA said.
Now Amazon has committed to doing less of that. "
less of that -> became stop :(
Re:Casio fx-CG50 calculator comes with Python built-in
on Monday October 30, 2023 @06:50AM (#38067814)
can confirm that - my child has one - no subscription needed for the calculator
Re:GraalOS: Containerless instant-on cloud functions for Java
on Saturday September 23, 2023 @05:25AM (#37621776)
https://www.graal.cloud/gcn/get-started/setting-up-desktop/ -> this details the actual steps to try this out.
Re:Human-Powered Air-Compressor and Energy Storage System
on Sunday July 16, 2023 @02:19PM (#36749988)
obligatory: fifteen million merits https://www.imdb.com/title/tt2089049/?ref_=ttep_ep2
Re:Google account deleted after 2 hours of Aurora
on Saturday May 27, 2023 @06:30PM (#36099044)
What is aurora?