lenz.io3w3 weeks ago

Five frontier LLMs disagree on 67% of 1k real-world fact-check claims (opens in new tab)

Covers Benchmark Illusion: Disagreement among LLMs and Its Scientific ConsequencesCovered by 4 sources See all sources covering this story including dbushell.com (blog), The New StackDiscussed on Hacker News

Feeds

✨ Discovered from this domain

Lenz — Audit-grade AI Fact-Checking API lenz.io

AI fact-checking API for products that can't afford to hallucinate. Audit-grade verdicts from a multi-model adversarial pipeline with a full citation trail.

Hacker News: Front Page hnrss.org

Hyundai buys Boston Dynamics, Atlas humanoid to be used at vehicle plant by 20281d1 day ago

How many of the 170k English words do you know?2d2 days ago

Project Valhalla Explained: How a Decade of Work Arrives in JDK 282d2 days ago

+25 more in the past week

Hacker News: Newest hnrss.org

Your brain was never designed for this much bad news11h11 hours ago

Finland's libraries are increasingly being valued not by how many books they lend, but how they help societies function16h16 hours ago

The Wholesale Plagiarism of Obscure Sorrows21h21 hours ago

+46 more in the past week

Hacker News: Best hnrss.org

Google hits 50% IPv66h6 hours ago

Your brain was never designed for this much bad news11h11 hours ago

Developers don't understand CORS (2019)13h13 hours ago

+2 more in the past day

Hacker News: Best hnrss.org

Google hits 50% IPv66h6 hours ago

Your brain was never designed for this much bad news11h11 hours ago

Developers don't understand CORS (2019)13h13 hours ago

+3 more in the past day

Hacker News: Newest hnrss.org

Google hits 50% IPv66h6 hours ago

Your brain was never designed for this much bad news11h11 hours ago

Developers don't understand CORS (2019)13h13 hours ago

+2 more in the past day

Hacker News: Newest hnrss.org

Google hits 50% IPv66h6 hours ago

Your brain was never designed for this much bad news11h11 hours ago

Developers don't understand CORS (2019)13h13 hours ago

+5 more in the past day

Hacker News: Newest hnrss.org

cssQuake1d1 day ago

I Stored a Website in a Favicon1d1 day ago

‘A man of great appetites’: what’s it like to be a dictator’s personal chef?1d1 day ago

+82 more in the past week

Hacker News: Newest hnrss.org

Anthropic to Require ID Verification for Certain Capabilities Starting July 82h2 hours ago

Beyond All Reason3h3 hours ago

Google hits 50% IPv66h6 hours ago

+9 more in the past day

Hacker News: Active hnrss.org

Anthropic to Require ID Verification for Certain Capabilities Starting July 82h2 hours ago

Beyond All Reason3h3 hours ago

The Case Against Geometric Algebra4h4 hours ago

+16 more in the past day

Hacker News: Active hnrss.org

Anthropic to Require ID Verification for Certain Capabilities Starting July 82h2 hours ago

Beyond All Reason3h3 hours ago

The Case Against Geometric Algebra4h4 hours ago

+18 more in the past day

Hacker News: Newest hnrss.org

Anthropic to Require ID Verification for Certain Capabilities Starting July 82h2 hours ago

Beyond All Reason3h3 hours ago

CTOs Agree: Cognitive Debt Is the New Technical Debt5h5 hours ago

+23 more in the past day

Hacker News: Front Page hnrss.org

Encryption, spyware, and now Mythos: History shows why cyber export control doesn’t work1d1 day ago

Lithuanian startup launches open-source network to detect Shahed-type drones1d1 day ago

16-year-old SATA II SSD survives 1 petabyte of writes — 25x more than the drive's endurance rating1d1 day ago

+169 more in the past week

Hacker News: Front Page hnrss.org

Who Actually Owns Your ATProto Identity? Hint: It's Probably Not You1h1 hour ago

David Ahl's Basic Computer Games Ported to C3h3 hours ago

The 100 Greatest Bird Names of All Time3h3 hours ago

+34 more in the past day

Hacker News news.ycombinator.com

Links for the intellectually curious, ranked by readers.

Former Olympian denies vandalising Washington Reflecting Pool after arrest33m33 minutes ago

Who Actually Owns Your ATProto Identity? Hint: It's Probably Not You1h1 hour ago

Anthropic to Require ID Verification for Certain Capabilities Starting July 82h2 hours ago

+41 more in the past day

progscrape progscrape.com

Robust Jobserver47m47 minutes ago

Performance improvements in libffi1h1 hour ago

Who Actually Owns Your ATProto Identity? Hint: It's Probably Not You1h1 hour ago

+62 more in the past day

Hacker News: Newest hnrss.org

Printing Gaussian Splats2h2 hours ago

Anthropic to Require ID Verification for Certain Capabilities Starting July 82h2 hours ago

David Ahl's Basic Computer Games Ported to C3h3 hours ago

+65 more in the past day

Hacker News: Newest hnrss.org

Show HN: I "built" Sana, a vector DB on object storage37m37 minutes ago

SpaceX handed lowest possible ESG rating by MSCI42m42 minutes ago

ZhuLinsen/daily_stock_analysis: LLM驱动的 A/H/美股智能分析器，多数据源行情 + 实时新闻 + Gemini 决策仪表盘 + 多渠道推送，零成本，纯白嫖，定时运行1h1 hour ago

+138 more in the past day

Hacker News: Newest hnrss.org

ZhuLinsen/daily_stock_analysis: LLM驱动的 A/H/美股智能分析器，多数据源行情 + 实时新闻 + Gemini 决策仪表盘 + 多渠道推送，零成本，纯白嫖，定时运行1h1 hour ago

Experiments in Sports Seismology for the World Cup1h1 hour ago

Kansas City's push for facial recognition on public buses starts privacy debate1h1 hour ago

+239 more in the past day

Hacker News: Newest hnrss.org

33 free financial calculators, all in the browser12m12 minutes ago

Xi: Microservices with built-in DI, compiled to a native binary12m12 minutes ago

The Scam Industry Is Booming [video][15 mins]22m22 minutes ago

+364 more in the past day

amerpie's Top Finds on Scour Today scour.ing

Ember, a native iOS Hacker News reader I built around accessibility22h22 hours ago

Toy Story has the right take on tech1d1 day ago

I hate compilers3d3 days ago

+8 more in the past week

blacknarwhal07's Top Finds on Scour Today scour.ing

AIPropel, AI-powered proposal generation for freelancers and agencies21h21 hours ago

Advice for Non-Pixel users?22h22 hours ago

Musician and YouTuber Hainbach on ‘Breath of the Wild’ and Swiss Army Knives23h23 hours ago

+48 more in the past week

micaleel's Top Finds on Scour Today scour.ing

Effective Use-Cases for LLMs10h10 hours ago

[NEW MODEL] SupraLabs started the Any2Any model family!11h11 hours ago

[Linkpost] How Transparent Is DiffusionGemma (and why it matters)19h19 hours ago

+4 more in the past day

Keyboard Shortcuts

Navigation

Next / previous post: j/k
Open post: oorEnter
Preview post: v

Post Actions

Love post: a
Like post: l
Dislike post: d
Undo reaction: u
Save / unsave: s

Recommendations

Add interest / feed: Enter
Not interested: x

Go to

Home: gh
Interests: gi
Feeds: gf
Likes: gl
History: gy
Changelog: gc
Settings: gs
Discover: gb
Search: /

Pagination

Next page: n
Previous page: p

General

Show this help: ?
Submit feedback: !
Close modal / unfocus: Esc

Press ? anytime to show this help