muelltonne@feddit.org to Technology@lemmy.worldEnglish · 11 days agoIt Only Takes A Handful Of Samples To Poison Any Size LLM, Anthropic Findshackaday.comexternal-linkmessage-square141linkfedilinkarrow-up1769arrow-down111
arrow-up1758arrow-down1external-linkIt Only Takes A Handful Of Samples To Poison Any Size LLM, Anthropic Findshackaday.commuelltonne@feddit.org to Technology@lemmy.worldEnglish · 11 days agomessage-square141linkfedilink
minus-squarejaykrown@lemmy.worldlinkfedilinkEnglisharrow-up2·10 days agoThat’s not how this works at all. The people training these models are fully aware of bad data. There are entire careers dedicated to preserving high quality data. GPT-4 is terrible compared to something like Gemini 3 Pro or Claude Opus 4.5.
That’s not how this works at all. The people training these models are fully aware of bad data. There are entire careers dedicated to preserving high quality data. GPT-4 is terrible compared to something like Gemini 3 Pro or Claude Opus 4.5.