K. Ramchandran

发表论文 38 篇 · 总被引 1569 次 · h-index 12

代表论文

• Why Do Multi-Agent LLM Systems Fail? (2025 · Neural Information Processing Systems · 被引 610)
• EmbedLLM: Learning Compact Representations of Large Language Models (2024 · International Conference on Learning Representations · 被引 50)
• Toward a Theory of Tokenization in LLMs (2024 · arXiv.org · 被引 40)
• Test Accuracy vs. Generalization Gap: Model Selection in NLP without Accessing Training or Testing Data (2023 · Knowledge Discovery and Data Mining · 被引 39)
• Transformers on Markov Data: Constant Depth Suffices (2024 · Neural Information Processing Systems · 被引 34)
• ProxySPEX: Inference-Efficient Interpretability via Sparse Feature Interactions in LLMs (2025 · Neural Information Processing Systems · 被引 17)