MAROKO133 Update ai: Nvidia Ridiculed for “Sloptracing” Feature That Uses AI to Yassify Vi

📌 MAROKO133 Eksklusif ai: Nvidia Ridiculed for “Sloptracing” Feature That Uses AI

Nvidia? The gaming GPU company?

On Monday, the multi-trillion dollar AI chipmaker unveiled its latest effort at weaving advances in AI into video games, and it immediately backfired.

The feature, DLSS 5, is supposed to be a souped-up version of the deep-learning upscaling tech Nvidia has offered since 2018. The company called it its “most significant breakthrough in computer graphics since the debut of real-time ray tracing” in that same year. But the reactions to demo footage shared has been overwhelmingly negative.

Gamers and developers fumed against the announcement, calling it “slop” and a “betrayal” of games’ artistic intent. Memes spread parodying the AI feature’s garish aesthetic, in which an original character or person is contrasted with a “DLSS 5” image that shows the subject in an unrecognizable style. Some even gave it a harsh nickname: “sloptracing,” a play on Nvidia’s ray tracing tech.

The reactions are warranted. Rather than just providing a little clarity to a fuzzy image, the feature looks more like a glorified Snapchat filter, varnishing the art style of your favorite games with an overwrought, generative AI finish. 

The effect is most noticeable when applied to faces. Iconic characters in the demo like Leon Kennedy from the Resident Evil franchise are, it’s no exaggeration to say, literally yassified.

According to Nvidia’s announcement, DLSS 5 “introduces a real-time neural rendering model that infuses pixels with photoreal lighting and materials.” It takes a “game’s color and motion vectors for each frame as input, and uses an AI model to infuse the scene with photoreal lighting and materials that are anchored to source 3D content.”

This AI model, it says, “is trained end to end to understand complex scene semantics such as characters, hair, fabric and translucent skin.”

Nvidia chief Jensen Huang was effusive about the tech’s implications, calling it gaming’s “GPT moment.”

“DLSS 5 is the GPT moment for graphics — blending hand-crafted rendering with generative AI to deliver a dramatic leap in visual realism while preserving the control artists need for creative expression,” he said in the announcement. 

It’s a little hard to buy Huang’s promise of preserving creative expression, however, when in all of the examples shared, DLSS 5 dramatically alters the aesthetic of the games. More than that, it exemplifies how generative AI uniformly reinforces bland aesthetic norms and defaults to gooner beauty standards. (Grace Ashcroft from the upcoming Resident Evil game gets hollower cheeks, stronger cheekbones, and poutier lips.) The games no longer look like games, but like any other clip spat out by a video generating model that gets shared in AI circles with a caption like “Hollywood is cooked.” 

Nvidia says DLSS 5 is arriving this fall — but, it seems, only to participating games that will include Resident Evil Requiem, Starfield, Hogwarts Legacy, and Assassin’s Creed Shadows. These are major titles, though, a show of how Nvidia says its feature is being supported by the industry’s biggest publishers and developers, like Capcom, Bethesda, Ubisoft, and Warner Bros Games.

More on AI: Unity Says It Has a New Product That Cooks Up Entire Games Using AI

The post Nvidia Ridiculed for “Sloptracing” Feature That Uses AI to Yassify Video Games in Real Time appeared first on Futurism.

🔗 Sumber: futurism.com

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

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