MAROKO133 Eksklusif ai: They Started an Entire College Dedicated to Resisting Cancel Cultu

📌 MAROKO133 Breaking ai: They Started an Entire College Dedicated to Resisting Can

In 2021, a crew of self-described free speech martyrs announced they were founding a university to save American higher education from the scourge of cancel culture. They called it the University of Austin, or UATX — and though it lacks accreditation, they boasted that it would be a place where controversial ideas could breathe free, and students wouldn’t live in fear of the woke mob.

Pretty much from the jump, critics of UATX suspected the project was less about “freedom of speech” and more about building a right-wing echo chamber for aggrieved libertarians. It operated out of a former retail store in downtown Austin, with funding from “anti-woke” philanthropists like Palantir co-founder Joseph Lonsdale.

Over the next few years, those critics would be proven correct in spectacular style. According to comprehensive reporting by Politico, in April of 2025 Lonsdale called an all-staff meeting that threw the entire project into a tailspin. Per the reporting, the billionaire told faculty and staff they must all subscribe to the “four principles of anti-communism, anti-socialism, identity politics, and anti-Islamism.” Anyone who didn’t toe the line would be frozen out.

In other words, the school supposedly founded to buck political litmus tests was issuing its own extremely stringent political litmus test — and threatening anyone who didn’t get on board with cancellation.

After that, everything changed. Staff and advisors began fleeing in droves. Students became disillusioned with the “university’s” mission. The founding president, Shakespeare scholar Pano Kanelos, stepped down from his role to take a largely ceremonial position as “Chancellor,” from which he would resign just a few months later.

One student told Politico: “I’ve never felt my speech was so chilled as it was in the classroom at UATX.”

Though a constitutional charter supposedly laid out a democratic process to resolve differences of opinion, these seemed to melt away at the request of figures like Lonsdale. The institution, as one staffer put it, had been “building the plane while we’re flying it.”

UATX, built to save students from the evils of cancel culture, has now become exactly what it claimed to oppose: an institution where the administration can end your academic career on a whim, where dissent from right-wing orthodoxy is not to be tolerated — and where the people most obsessed with cancel culture can finally do the cancelling.

More on education: Gen Z Arriving at College Unable to Read

The post They Started an Entire College Dedicated to Resisting Cancel Culture, and Then the Funniest Possible Thing Happened appeared first on Futurism.

🔗 Sumber: futurism.com

📌 MAROKO133 Eksklusif ai: Adobe Research Unlocking Long-Term Memory in Video World

Video world models, which predict future frames conditioned on actions, hold immense promise for artificial intelligence, enabling agents to plan and reason in dynamic environments. Recent advancements, particularly with video diffusion models, have shown impressive capabilities in generating realistic future sequences. However, a significant bottleneck remains: maintaining long-term memory. Current models struggle to remember events and states from far in the past due to the high computational cost associated with processing extended sequences using traditional attention layers. This limits their ability to perform complex tasks requiring sustained understanding of a scene.

A new paper, “Long-Context State-Space Video World Models” by researchers from Stanford University, Princeton University, and Adobe Research, proposes an innovative solution to this challenge. They introduce a novel architecture that leverages State-Space Models (SSMs) to extend temporal memory without sacrificing computational efficiency.

The core problem lies in the quadratic computational complexity of attention mechanisms with respect to sequence length. As the video context grows, the resources required for attention layers explode, making long-term memory impractical for real-world applications. This means that after a certain number of frames, the model effectively “forgets” earlier events, hindering its performance on tasks that demand long-range coherence or reasoning over extended periods.

The authors’ key insight is to leverage the inherent strengths of State-Space Models (SSMs) for causal sequence modeling. Unlike previous attempts that retrofitted SSMs for non-causal vision tasks, this work fully exploits their advantages in processing sequences efficiently.

The proposed Long-Context State-Space Video World Model (LSSVWM) incorporates several crucial design choices:

1. Block-wise SSM Scanning Scheme: This is central to their design. Instead of processing the entire video sequence with a single SSM scan, they employ a block-wise scheme. This strategically trades off some spatial consistency (within a block) for significantly extended temporal memory. By breaking down the long sequence into manageable blocks, they can maintain a compressed “state” that carries information across blocks, effectively extending the model’s memory horizon.
2. Dense Local Attention: To compensate for the potential loss of spatial coherence introduced by the block-wise SSM scanning, the model incorporates dense local attention. This ensures that consecutive frames within and across blocks maintain strong relationships, preserving the fine-grained details and consistency necessary for realistic video generation. This dual approach of global (SSM) and local (attention) processing allows them to achieve both long-term memory and local fidelity.

The paper also introduces two key training strategies to further improve long-context performance:

• Diffusion Forcing: This technique encourages the model to generate frames conditioned on a prefix of the input, effectively forcing it to learn to maintain consistency over longer durations. By sometimes not sampling a prefix and keeping all tokens noised, the training becomes equivalent to diffusion forcing, which is highlighted as a special case of long-context training where the prefix length is zero. This pushes the model to generate coherent sequences even from minimal initial context.
• Frame Local Attention: For faster training and sampling, the authors implemented a “frame local attention” mechanism. This utilizes FlexAttention to achieve significant speedups compared to a fully causal mask. By grouping frames into chunks (e.g., chunks of 5 with a frame window size of 10), frames within a chunk maintain bidirectionality while also attending to frames in the previous chunk. This allows for an effective receptive field while optimizing computational load.

The researchers evaluated their LSSVWM on challenging datasets, including Memory Maze and Minecraft, which are specifically designed to test long-term memory capabilities through spatial retrieval and reasoning tasks.

The experiments demonstrate that their approach substantially surpasses baselines in preserving long-range memory. Qualitative results, as shown in supplementary figures (e.g., S1, S2, S3), illustrate that LSSVWM can generate more coherent and accurate sequences over extended periods compared to models relying solely on causal attention or even Mamba2 without frame local attention. For instance, on reasoning tasks for the maze dataset, their model maintains better consistency and accuracy over long horizons. Similarly, for retrieval tasks, LSSVWM shows improved ability to recall and utilize information from distant past frames. Crucially, these improvements are achieved while maintaining practical inference speeds, making the models suitable for interactive applications.

The Paper Long-Context State-Space Video World Models is on arXiv

The post Adobe Research Unlocking Long-Term Memory in Video World Models with State-Space Models first appeared on Synced.

🔗 Sumber: syncedreview.com