WeSearch

Tell HN: Gemini 3.5 Flash breaks in stupid ways

·1 min read · 0 reactions · 0 comments · 17 views
#technology#ai#modeling
⚡ TL;DR · AI summary

Users are experiencing issues with the Gemini 3.5 Flash model, specifically in its grading functionality. When a 'Grading criteria' text is added, the model tends to underperform by collapsing scores towards the center of the scale. This behavior raises concerns about the reliability of current state-of-the-art models despite their advancements in other areas.

Key facts
Original article
Ycombinator
Read full at Ycombinator →
Opening excerpt (first ~120 words) tap to expand

I thought I was going crazy, trying to use Gemini 3.5 Flash to rate some answers, but it kept giving 7 instead of 10 for correct answers.Apparently once you add a "Grading criteria" text, the model collapses into a "compressed toward the center of the scale" hallucination (or training set overfitting).Someone on X asked me to try to reproduce it, and I actually got it on the first try on their Gemini Chat: https://x.com/XCSme/status/2057613611959279988I am not sure what to make of this (or most SOTA) models. They got a lot smarter with coding and tool usage, but a lot dumber in other ways...

Excerpt limited to ~120 words for fair-use compliance. The full article is at Ycombinator.

Anonymous · no account needed
Share 𝕏 Facebook Reddit LinkedIn Threads WhatsApp Bluesky Mastodon Email

Discussion

0 comments

More from Ycombinator