Machine unlearning is a promising solution for efficiently removing specific knowledge by post hoc modifying models.

Large language models (LLMs) inevitably memorize sensitive, copyrighted, and harmful knowledge from the training corpus; therefore, it is crucial to erase this knowledge from the models. Machine unlearning is a promising solution for efficiently removing specific knowledge by post hoc modifying models. In this paper, we propose a Real-World Knowledge Unlearning benchmark (RWKU) for LLM unlearning. RWKU is designed based on the following three key factors: (1) For the task setting, we consider a more practical and challenging unlearning setting, where neither the forget corpus nor the retain corpus is accessible. (2) For the knowledge source, we choose 200 real-world famous people as the unlearning targets and show that such popular knowledge is widely present in various LLMs. (3) For the evaluation framework, we design the forget set and the retain set to evaluate the model's capabilities across various real-world applications. Regarding the forget set, we provide four four membership inference attack (MIA) methods and nine kinds of adversarial attack probes to rigorously test unlearning efficacy. Regarding the retain set, we assess locality and utility in terms of neighbor perturbation, general ability, reasoning ability, truthfulness, factuality, and fluency. We conduct extensive experiments across two unlearning scenarios, two models and six baseline methods and obtain some meaningful findings. We release our benchmark and code publicly at http://rwku-bench.github.io for future work.

A new sci-fit movie has overtaken The Shawshank Redemption as the number one highest-rated film on IMDB - and it hasn't even been released. The Shawshank Redemption has held the number one best film spot on IMDB since 2008 when it knocked The Godfather out of its long-standing ranking of nearly 20 years. The 1994 flick has lost its crown Dune: Part Two, which has been deemed'visually thrilling' and'narratively epic' by critics days before its official release on March 1. The Warner Bros movie has a 9.4 rating on IMDB from viewers who went to an early screening of the Sci-Fi epic on Sunday, along with a 97 percent on Rotten Tomatoes. Dune: Part Two is the second park in the film adaptation of Frank Herbert's first book of the same name Dune: Part Two's sudden rise in popularity is in direct contrast to The Shawshank Redemption, which was a slow burn for viewers after it tanked at the box office.

Professional movie critics aren't enjoying'Indiana Jones and the Dial of Destiny,' the hotly anticipated, and allegedly final, adventure for Harrison Ford as the whip-cracking archeologist. Pre-release reviews for the picture have lead to a'rotten' 50 percent score at Rotten Tomatoes, based on 46 reviews. And the aggregation site Metacritic gives the new Indy a score of 52/100, based on 24 reviews. But those failing grades could mean that'Indy 5' is shaping up to be a runaway summer sensation -- according to researchers at University of California Davis. Pre-release reviews for'Indiana Jones and the Dial of Destiny' have lead to a'rotten' score of 50% at Rotten Tomatoes and a 52/100 at Metacritic.

The viral Donald Trump arrest and prison escape photos are hilarious, and were based on a famous movie! Take a look for yourself, and learn about the story behind these viral photos! BREAKING: Donald Trump was just arrested by New York law enforcement. Recommended: Did Simon Cowell Just Die In A Car Accident?! The truth is these viral Donald Trump arrest photos are fake, and were generated by artificial intelligence, and is really based on a famous Hollywood movie! First, let me just confirmed that all those viral photos of police officers trying to arrest Donald Trump are fake, and were generated by artificial intelligence.

Watching meaningful movies such as Up or Slumdog Millionaire can help us feel more prepared to deal with challenges and want to be a better person, a new study shows. Ohio State University researchers created two lists of films made after 1985 with high viewer ratings - one with meaningful and one with less meaningful movies. They then had 1,098 volunteers watch either the meaningful or less meaningful list of movies, before filling in a survey on their thoughts and reaction to the films. Watching meaningful films - those that we find moving and poignant - can make us feel more prepared to deal with life's challenges, the authors found. The team say this could explain why people turn to movies that make them both sad and happy and explore difficult subjects they may not always find uplifting.

For decades, cable has roped in millions of customers like me with the promise of hundreds of channels and thousands of shows. But in my 15-plus years as a subscriber, there's one thing I've watched most: my bill. Every month I pay it, and every month I think of cutting the cord. The reason is that there's never anything good on--unless you're a fan of The Shawshank Redemption (which is probably on two channels at once), or one of the 19 shows based on storage units. Years of this feeling has brought people like me to a slow boil and caused them to pull the plug on their pay television.