Πηγαίνετε εκτός σύνδεσης με την εφαρμογή Player FM !
[QA] Poser: Unmasking Alignment Faking LLMs by Manipulating Their Internals
Manage episode 417498419 series 3524393
Large Language Models (LLMs) can deceive as 'alignment fakers.' A benchmark with 324 LLM pairs is introduced to detect misbehaving models, achieving 98% accuracy with a specific strategy.
https://arxiv.org/abs//2405.05466
YouTube: https://www.youtube.com/@ArxivPapers
TikTok: https://www.tiktok.com/@arxiv_papers
Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016
Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers
--- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/support
1117 επεισόδια
Manage episode 417498419 series 3524393
Large Language Models (LLMs) can deceive as 'alignment fakers.' A benchmark with 324 LLM pairs is introduced to detect misbehaving models, achieving 98% accuracy with a specific strategy.
https://arxiv.org/abs//2405.05466
YouTube: https://www.youtube.com/@ArxivPapers
TikTok: https://www.tiktok.com/@arxiv_papers
Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016
Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers
--- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/support
1117 επεισόδια
Όλα τα επεισόδια
×Καλώς ήλθατε στο Player FM!
Το FM Player σαρώνει τον ιστό για podcasts υψηλής ποιότητας για να απολαύσετε αυτή τη στιγμή. Είναι η καλύτερη εφαρμογή podcast και λειτουργεί σε Android, iPhone και στον ιστό. Εγγραφή για συγχρονισμό συνδρομών σε όλες τις συσκευές.