RLHF getting out of hand, netizen mocks current state
omooretweets · x · 2026-08-24
Shares a link mocking the current state of RLHF (Reinforcement Learning from Human Feedback), suggesting it has become excessive or uncontrolled.
More from Fun
- Grok flatters too: 'draw me from my enemies' view' goes comically sideways — repligate · 2026-08-24
- Autonomous 5v5 humanoid soccer kicks off; fall recovery improves but possession still hard — rohanpaul_ai · 2026-08-24
- Dadabots proposes '3-era' meme: AI is dead, long live engineering — teropa · 2026-08-24
- Tiangong robot wins race with hilarious running gait — gaganghotra_ · 2026-08-24
- From Jurassic Park's Dinosaur Input Device to Hand-Tracked Blender in Vision Pro — bilawalsidhu · 2026-08-24
- Humanoid design has reached the "Skyrim character creator" stage — carlosdponx · 2026-08-24