3 accepted papers
In retrieval augmented generation (RAG) systems, each individual component—the LLM, embedder, and corpus—could introduce biases in the form of skews towards certain genders or political leanings. In this work, we study the conflict between biases of each component and their relationship to the overa…
Reasoning methods that adaptively allocate test-time compute have advanced LLM performance on easy to verify domains such as math and code. In this work, we study how to utilize this approach to train models that exhibit a degree of robustness to safety vulnerabilities, and show that doing so can pr…
Estimating the out-of-distribution performance in regimes where labels are scarce is critical to safely deploy foundation models. Recently, it was shown that ensembles of neural networks observe the phenomena "agreement-on-the-line", which can be leveraged to reliably predict OOD performance without…