Ari Holtzman
发表论文 6 篇 · 总被引 38 次 · h-index 3
代表论文
- LLM Probability Concentration: How Alignment Shrinks the Generative Horizon (2025 · 被引 23)
- LL3M: Large Language 3D Modelers (2025 · arXiv.org · 被引 17)
- Mapping Overlaps in Benchmarks through Perplexity in the Wild (2025 · arXiv.org · 被引 5)
- Dynamic Weight Grafting: Localizing Finetuned Factual Knowledge in Transformers (2025 · 被引 1)
- Multiple Streams of Relation Extraction: Enriching and Recalling in Transformers (2025 · arXiv.org)
- Interpretation Errors: Extracting Functionality from Generative Models of Language by Understanding Them Better (2023)