60 citations · 302 across the 41 of their papers we have counts for
6 papers · 1 filter
PRISM Edit: One Vector for All Temporal Answers
Chen Huang, Qi Zheng, Ruiqin Zheng +2
Model editing keeps large language models (LLMs) up to date without retraining, but temporal facts expose a limitation of the prevailing locate-and-edit paradigm: an update is not…
Certifiably Robust RAG against Retrieval Corruption
Chong Xiang, Tong Wu, Zexuan Zhong +3
Retrieval-augmented generation (RAG) is susceptible to retrieval corruption attacks, where malicious passages injected into retrieval results can lead to inaccurate model responses…
Foundational Challenges in Assuring Alignment and Safety of Large Language Models
Usman Anwar, Abulhair Saparov, Javier Rando +39
This work identifies 18 foundational challenges in assuring the alignment and safety of large language models (LLMs). These challenges are organized into three different categories…
Interpretability Illusions in the Generalization of Simplified Models
Dan Friedman, Andrew Lampinen, Lucas Dixon +2
A common method to study deep learning systems is to use simplified model representations--for example, using singular value decomposition to visualize the model's hidden states in…
Learning Transformer Programs
Dan Friedman, Alexander Wettig, Danqi Chen
Recent research in mechanistic interpretability has attempted to reverse-engineer Transformer models by carefully inspecting network weights and activations. However, these approac…
Fine-Tuning Language Models with Just Forward Passes
Sadhika Malladi, Tianyu Gao, Eshaan Nichani +4
Fine-tuning language models (LMs) has yielded success on diverse downstream tasks, but as LMs grow in size, backpropagation requires a prohibitively large amount of memory. Zeroth-…