Hackernews posts about Ben

1. Meta got caught gaming AI benchmarks (www.theverge.com)
   347 points by pseudolus 13 days ago | 161 comments
2. Installing air filters in classrooms has surprisingly large educational benefits (2020) (www.vox.com)
   308 points by tangjurine 21 days ago | 142 comments
3. Benchmarking LLM social skills with an elimination game (github.com)
   194 points by colonCapitalDee 17 days ago | 60 comments
4. Benn Jordan's AI poison pill and the weird world of adversarial noise (cdm.link)
   132 points by glitcher 6 days ago | 201 comments
5. Show HN: LocalScore – Local LLM Benchmark (www.localscore.ai)
   124 points by sipjca 18 days ago | 25 comments
6. LLM Benchmark for 'Longform Creative Writing' (eqbench.com)
   96 points by vitorgrs 11 days ago | 88 comments
7. Show HN: Benchi – A benchmarking tool written in Go (github.com)
   58 points by lmazgon 18 days ago | 7 comments
8. Which countries would benefit most from an American brain drain? (www.economist.com)
   44 points by rustoo 9 days ago | 78 comments
9. OpenAI wants to bend copyright rules. Study suggests it isn't waiting (www.theregister.com)
   19 points by pseudolus 17 days ago | 7 comments
10. New open-source benchmark for real-time analytics applications (github.com)
    16 points by thenoahhein 25 days ago | 4 comments
11. Benefits of Apache Iceberg for geospatial data analysis (wherobots.com)
    16 points by MrPowers 9 days ago | 1 comments
12. Ask HN: Do You Take Protein Suppliments? Any Noticeable Benefits?
    15 points by samrohn 19 days ago | 13 comments
13. Therapy chatbot trial yields mental health benefits (home.dartmouth.edu)
    14 points by geox 23 days ago | 4 comments
14. Vending-Bench: A Benchmark for Long-Term Coherence of Autonomous Agents (arxiv.org)
    14 points by distalx 1 day ago | discuss
15. Medical Benchmarks and the Myth of the Universal Patient (www.newyorker.com)
    12 points by pseudolus 25 days ago | 3 comments
16. NIST's Curved Neutron Beams Could Deliver Benefits Straight to Industry (www.nist.gov)
    11 points by rbanffy 4 days ago | discuss
17. DeepSeek-V3-0324 Crushes GPT-4.5 in Math and Code Benchmarks at 1/277 the Cost (api-docs.deepseek.com)
    10 points by kong75 27 days ago | 2 comments
18. We need a better term for GenAI output – "slop" is too benign (www.rockpapershotgun.com)
    10 points by fjghajkhdfgjlk 19 days ago | discuss
19. Homeownership is 'an investment,' lawmaker says. Fewer adults are benefitting (www.cnbc.com)
    10 points by donsupreme 21 days ago | discuss
20. #1 open-source agent on SWE-Bench Verified by combining Claude 3.7 and O1 (www.augmentcode.com)
    9 points by emmabotbot 20 days ago | 3 comments
21. LocalScore: A Local LLM Benchmark (www.localscore.ai)
    9 points by jborichevskiy 17 days ago | 2 comments
22. This Bench Does Not Exist (doesnotexist.openbenches.org)
    8 points by thinkingemote 13 days ago | 6 comments
23. DeepSeek V3 0324 outpaces GPT 4.5 and Claude 3.7 in coding, other benchmarks (huggingface.co)
    7 points by xianshou 25 days ago | discuss
24. Why Software Consultants Benefit from Liberal Arts Education (spin.atomicobject.com)
    6 points by philk10 14 days ago | 3 comments
25. NPB-Rust: NAS Parallel Benchmarks in Rust (arxiv.org)
    6 points by aw1621107 7 days ago | 1 comments
26. Show HN: CVE-Bench, the first LLM benchmark using real-world web vulnerabilities (github.com)
    6 points by jbenn 20 days ago | 1 comments
27. The Curve Is Bending (grantslatton.com)
    6 points by delifue 14 days ago | discuss
28. No Man's Sky's newest update adds bones and fossils beneath the Earth (www.polygon.com)
    6 points by amichail 26 days ago | discuss
29. Vending-Bench: Testing long-term coherence in agents (andonlabs.com)
    5 points by tosh 2 days ago | 2 comments
30. Don't overlook the many benefits of plastics (www.economist.com)
    5 points by amadeuspagel 1 day ago | 1 comments