Scoring generations of the same news against each other to filter AI slop
ivan_bezdomny · x · 2026-08-25
Asked what dataset he started with to remove AI slop, ivanbezdomny explains his approach: he scores different generations of the same news against each other to measure quality (noting that SFT doesn't really work anyway), and uses real tweets to count word and phrase frequencies as a proxy for quality writing on the same topic. A longer post is in the works.
More from Research
- CFP: UIST 2026 Workshop on Augmenting Human Dexterity — plopesresearch · 2026-08-27
- 60fps depth estimation achieved targeting 20ms end-to-end latency — yacineMTB · 2026-08-27
- Q-Planning: robot policies self-improve from 25% to 80% success, no new demos — rbhar90 · 2026-08-27
- CrysVCD enforces chemistry constraints in crystal generation — bravo_abad · 2026-08-27
- Continual Learning is AI's Holy Grail: Fluid intelligence will enable the next capability leap — daniel_mac8 · 2026-08-27
- How to achieve same intelligence with less computation? Exploring 7 technical paths — prateekj · 2026-08-27