lamini-ai/Lamini-Memory-Tuning
Banishing LLM Hallucinations Requires Rethinking Generalization
2 best points · 0 comments · 2 Hacker News posts · 277 GitHub stars · View on GitHub
Hacker News discussions
- Lamini Memory Tuning: near-perfect fact recall via 1M-way MoE [pdf] (2 points, 0 comments, by Bluestein, 2024-06-15)
- Banishing LLM hallucinations requires rethinking generalization [pdf] (1 points, 0 comments, by itunpredictable, 2024-06-13)