▶ ai·
items10 latest
▶ ai·
Alignment pretraining: AI discourse creates self-fulfilling (mis)alignment
▶ ai·
Self-Distillation Enables Continual Learning [pdf]
▶ ai·
Δ-Mem: Efficient Online Memory for Large Language Models
▶ ai·
New arXiv policy: 1-year ban for hallucinated references
▶ ai·
Deterministic Fully-Static Whole-Binary Translation Without Heuristics
▶ ai·
Beyond Semantic Similarity
▶ ai·
LLMorphism: When humans come to see themselves as language models
▶ ai·
LLMs Corrupt Your Documents When You Delegate
▶ ai·