📌 TOPINDIATOURS Update ai: NASA’s Moon Spacesuits Are Plagued With Issues Wajib Ba

NASA is gearing up for the first crewed journey to the Moon in over half a century, a mission that could launch as soon as two weeks from now.

And next year, the agency will finally attempt to return astronauts to the lunar surface itself as part of its Artemis 3 mission, which will dramatically increase the already considerable stakes.

Particularly when it comes to stepping out of the spacecraft — the agency has yet to pick between Blue Origin and SpaceX’s offerings in that regard — staying protected from the extreme temperature swings, space radiation, and lack of atmosphere is extremely challenging.

That’s not to mention the physical limitations of an extremely bulky spacesuit, which could physically tax astronauts even more than stepping outside of the International Space Station during a spacewalk.

As Ars Technica reports, former NASA astronaut and microbiologist Kate Rubins, who retired last year and has logged 300 days in space, recently voiced her concerns over the Moon suit that private space company Axiom Space has been developing for NASA as part of a $228 million contract.

“What I think we have on the Moon that we don’t really have on the space station that I want people to recognize is an extreme physical stress,” she said during a recent meeting of the National Academies of Sciences, Engineering, and Medicine.

Besides not getting any sleep, Rubin warned that people will be “in these suits for eight or nine hours” and doing extravehicular activities (EVAs) “every day.”

Compared to the suits NASA astronauts wore during the Apollo missions, the Axiom Space suit is considerably heavier. While a sixth of gravity will greatly alleviate some of that heft, they still weigh in at 300 pounds. At the same time, Moon walkers will enjoy greatly enhanced flexibility, allowing them to kneel down to pick up objects, for instance.

“I think the suits are better than Apollo, but I don’t think they are great right now,” Rubin warned, noting “flexibility issues” and the reality that “people are going to be falling over.”

In remarks directly to Ars, Rubin elaborated, emphasizing that the suits are “definitely much better than Apollo,” but remain “still quite heavy.”

Even something as simple as getting back up after a fall — as demonstrated by the many Apollo astronauts who took a tumble while on the Moon — involves a type of “jumping pushup,” as Rubins told Ars, which is a “non-trivial” and “risky maneuver.”

Not everybody is as concerned about the Axiom Space suit. Current NASA astronaut and physician Mike Barratt argued in remarks during the committee meeting that the “suit is getting there,” pointing out that “we’ve got 700 hours of pressurized experience in it right now.”

“Bending down in the suit is really not too bad at all,” he added.

NASA still plans to conduct plenty of tests involving the suit, including parabolic flight, which can simulate the partial gravity of the Moon’s surface. The agency has already put the suit through its paces underwater at NASA’s Neutral Buoyancy Lab.

The agency has until sometime next year to finalize the design for its long-awaited Artemis 3 mission to the lunar surface. At the same time, NASA still has plenty of decisions to make, including how to get down to the lunar surface in the first place.

More on Artemis: Experts Warn That There’s Something Wrong With the Moon Rocket NASA Is About to Launch With Astronauts Aboard

The post NASA’s Moon Spacesuits Are Plagued With Issues appeared first on Futurism.

🔗 Sumber: futurism.com

📌 TOPINDIATOURS Eksklusif ai: Adobe Research Unlocking Long-Term Memory in Video W

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