影响力指数

28.75/100

前 25.9%

全站排名 #16,644

发表论文5 篇

平均评分5.4

年均产出2.5 篇/年

Vikrant Varma

Researcher@Google·英国·OpenReview

MONA: Myopic Optimization with Non-myopic Approval Can Mitigate Multi-step Reward Hacking

ICML 2025Poster

Jumping Ahead: Improving Reconstruction Fidelity with JumpReLU Sparse Autoencoders

ICLR 2025Rejected

Improving Sparse Decomposition of Language Model Activations with Gated Sparse Autoencoders

NeurIPS 2024Poster

Explaining grokking through circuit efficiency

ICLR 2024Rejected

Challenges with unsupervised LLM knowledge discovery

NeurIPS 2024Rejected

合作者 (17)

Sebastian Farquhar

Senthooran Rajamanoharan