From the 1 of 7 linked papers with an AI index.
7 papers
GPT-Red: Automated Red Teaming via Self-Play at Scale
Eric Wallace, Christopher A. Choquette-Choo, Nikhil Kandpal +15
The paper presents GPT-Red, an automated red‑teaming system that uses self‑play to generate novel prompt‑injection attacks against large language models and improve their robustnes…
IH-Challenge: A Training Dataset to Improve Instruction Hierarchy on Frontier LLMs
Chuan Guo, Juan Felipe Ceron Uribe, Sicheng Zhu +10
Instruction hierarchy (IH) defines how LLMs prioritize system, developer, user, and tool instructions under conflict, providing a concrete, trust-ordered policy for resolving instr…
Enhancing Training Data Attribution with Representational Optimization
Weiwei Sun, Haokun Liu, Nikhil Kandpal +2
Training data attribution (TDA) methods aim to measure how training data impacts a model's predictions. While gradient-based attribution methods, such as influence functions, offer…
Efficient Model Development through Fine-tuning Transfer
Pin-Jie Lin, Rishab Balasubramanian, Fengyuan Liu +2
Modern LLMs struggle with efficient updates, as each new pretrained model version requires repeating expensive alignment processes. This challenge also applies to domain- or langua…
The Common Pile v0.1: An 8TB Dataset of Public Domain and Openly Licensed Text
Nikhil Kandpal, Brian Lester, Colin Raffel +24
Large language models (LLMs) are typically trained on enormous quantities of unlicensed text, a practice that has led to scrutiny due to possible intellectual property infringement…
Position: The Most Expensive Part of an LLM should be its Training Data
Nikhil Kandpal, Colin Raffel
Training a state-of-the-art Large Language Model (LLM) is an increasingly expensive endeavor due to growing computational, hardware, energy, and engineering demands. Yet, an often-…