pith. sign in

Vladimir Mikulik

Identifiers

  • name variant Vladimir Mikulik 0.60 · backfill

Papers (15)

  1. Human-AI Complementarity: A Goal for Amplified Oversight cs.AI · 2025 · author #6
  2. An Approach to Technical AGI Safety and Security cs.AI · 2025 · author #23
  3. Gemini: A Family of Highly Capable Multimodal Models cs.CL · 2023 · author #1222
  4. Challenges with unsupervised LLM knowledge discovery cs.LG · 2023 · author #5
  5. The Hydra Effect: Emergent Self-repair in Language Model Computations cs.LG · 2023 · author #4
  6. Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla cs.LG · 2023 · author #7
  7. Tracr: Compiled Transformers as a Laboratory for Interpretability cs.LG · 2023 · author #6
  8. Teaching language models to support answers with verified quotes cs.CL · 2022 · author #3
  9. Scaling Language Models: Methods, Analysis & Insights from Training Gopher cs.CL · 2021 · author #58
  10. Alignment of Language Agents cs.AI · 2021 · author #5
  11. Causal Analysis of Agent Behavior for AI Safety cs.AI · 2021 · author #6
  12. Algorithms for Causal Reasoning in Probability Trees cs.AI · 2020 · author #4
  13. Meta-trained agents implement Bayes-optimal agents cs.AI · 2020 · author #1
  14. Neural networks are a priori biased towards Boolean functions with low entropy cs.LG · 2019 · author #5
  15. Risks from Learned Optimization in Advanced Machine Learning Systems cs.AI · 2019 · author #3

Mentions

  • 2504.01849 #23 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 2312.10029 #5 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 2301.05062 #6 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 2307.15771 #4 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 2307.09458 #7 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 1909.11522 #5 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 2112.11446 #58 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 1906.01820 #3 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 2103.14659 #5 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 2103.03938 #6 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 2010.12237 #4 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 2010.11223 #1 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 2510.26518 #6 · arxiv_oai · confidence 0.70 Vladimir Mikulik
  • 2203.11147 #3 · arxiv_oai · confidence 0.70 Vladimir Mikulik

Frequent Coauthors