VDC: Versatile Data Cleanser for Detecting Dirty Samples via Visual-Linguistic Inconsistency Paper • 2309.16211 • Published Sep 28, 2023
BoT: Breaking Long Thought Processes of o1-like Large Language Models through Backdoor Attack Paper • 2502.12202 • Published 24 days ago
BackdoorBench: A Comprehensive Benchmark of Backdoor Learning Paper • 2206.12654 • Published Jun 25, 2022
BoT: Break CoT Collection Break long thought processes of large reasoning models, including DeepSeek-R1, QwQ, Marco-o1 • 11 items • Updated 6 days ago
BoT: Break CoT Collection Break long thought processes of large reasoning models, including DeepSeek-R1, QwQ, Marco-o1 • 11 items • Updated 6 days ago