Publications
Explore a selection of our recent research on some of the most complex and interesting challenges in AI.
259 publications
-
- 20 February 2025
- Delta Variances
-
- 12 February 2025
- Poly-Autoregressive Prediction for Interaction Modeling
-
- 11 February 2025
- Scaling Pre-training to One Hundred Billion Data for Vision Language Models
-
- 6 February 2025
- Automated Discovery of Interpretable Cognitive Programs underlying Reward-guided behavior
-
- 2 February 2025
- Decoding-based Regression
-
- 22 January 2025
- MONA: Myopic Optimization with Non-myopic Approval Can Mitigate Multi-step Reward Hacking
-
- 22 January 2025
- Are vision-language models shape or texture biased and can we steer them?
-
- 17 January 2025
- Evolving Deeper LLM Thinking
-
- 8 January 2025
- Foundations of Algorithmic Thermodynamics
-
- 31 December 2024
- Exposing Limitations of Language Model Agents in Sequential-Task Compositions on the Web
-
- 26 December 2024
- A theory of appropriateness with applications to generative artificial intelligence
-
- 23 December 2024
- Deliberation in Latent Space via Differentiable Cache Augmentation
-
- 10 December 2024
- What type of inference is planning?
-
- 10 December 2024
- Machine Unlearning Doesn’t Do What You Think: Lessons for Generative AI Policy, Research, and Practice
-
- 6 December 2024
- Exponential Speedups by Rerooting Levin Tree Search
-
- 4 December 2024
- Mastering Board Games by External and Internal Planning with Language Models
-
- 26 November 2024
- How Well Do Large Language Models Perform Latent Multi-Hop Reasoning without Exploiting Shortcuts?
-
- 22 November 2024
- Understanding LLM Embeddings for Regression
-
- 18 October 2024
- AI can help humans find common ground in democratic deliberation
-
- 17 October 2024
- Prompting Considered Harmful
-
- 14 October 2024
- Predicting from Strings: Language Model Embeddings for Bayesian Optimization
-
- 9 October 2024
- Diffusion model predictive control
-
- 26 September 2024
- Geometric-Averaged Preference Optimization for Soft Preference Labels
-
- 21 September 2024
- Learned feature representations are biased by complexity, learning order, position, and more
-
- 19 September 2024
- Michelangelo: Long Context Evaluations Beyond Haystacks via Latent Structure Queries
-
- 7 September 2024
- Modeling the Arrows of Time with Causal Multibaker Maps
-
- 21 August 2024
- The Vizier Gaussian Process Bandit Algorithm
-
- 20 August 2024
- Swim till you sink: Computing the limit of a game
-
- 13 August 2024
- The Probabilities Also Matter: A More Faithful Metric for Faithfulness of Free-Text Explanations in Large Language Models
-
- 7 August 2024
- Achieving Human Level Competitive Robot Table Tennis