MAROKO133 Update ai: Researchers unveil magnet that slashes alloy heat treatment time and

📌 MAROKO133 Hot ai: Researchers unveil magnet that slashes alloy heat treatment ti

Steel parts and aluminum cans could soon be made with far less energy, thanks to the University of Florida.

In a collaboration backed by nearly $11 million in federal funding, UF scientists have developed a first-of-its-kind superconducting magnet. The device could transform metal production and boost U.S. leadership in alloy manufacturing.

The technology targets the steel industry.

By combining magnetic fields with high-temperature heat treatment, it could drastically reduce processing time and energy use.

Fast Processing

Funded by the U.S. Department of Energy’s Advanced Manufacturing Office, the project uses Induction-Coupled Thermomagnetic Processing, or ITMP. It integrates magnetic fields with thermal processing for unprecedented efficiency.

“This revolutionary technology has the potential to substantially reduce the cost and energy use of heat treatments in the steel industry, and we are excited to help pave the way for its adoption in industry,” said Michael Tonks, Ph.D., UF’s interim chair of Materials Science and Engineering.

The magnet is custom-built and paired with a cylinder induction furnace. Together, they sit atop a six-foot-high platform. The $6 million prototype can process steel samples up to five inches in diameter, making it rare for academic research.

Yang Yang, Ph.D., UF materials science research faculty member, said ITMP could reduce steel processing time by up to 80 percent. “Thermomagnetic processing changes a material’s phase stability and kinetic properties, accelerating carbon diffusion in steel,” Yang said. “Traditional furnace system modifies the driving forces for steel phase changes. What normally takes eight hours can be done in just a few minutes. “The magnetic field acts as an external driving force to make atoms diffuse faster,” Yang explained.

Unlike conventional energy sources such as electricity or natural gas, ITMP uses volumetric induction heating and high-static magnetic fields. The result is lower energy consumption and reduced operational costs.

Industrial impact

At Oak Ridge National Laboratory, collaborators praised the UF prototype. “This could significantly advance U.S. manufacturing and process efficiency for heat treatment of materials such as metal alloys of steel or aluminum,” said Michael Kesler, Ph.D., ORNL research scientist and lead collaborator.

Researchers said ITMP could support a more reliable energy grid and improve industrial electrification. UF scientists also note that it could cut carbon emissions, paving the way for cleaner manufacturing.

The magnet now resides in UF’s Powell Family Structures and Materials Laboratory on the East Campus. It will be officially unveiled in December with representatives from national labs, industry, and academia. Engineering students will have opportunities for research and hands-on learning.

The project is still in a pilot phase and requires further testing. However, UF researchers are optimistic about future industry adoption within five to ten years.

The initiative is part of a $187 million DOE program to strengthen U.S. manufacturing competitiveness. If successful, the technology could redefine how steel, aluminum, and other metal alloys are produced worldwide.

🔗 Sumber: interestingengineering.com

📌 MAROKO133 Breaking ai: MIT Researchers Unveil “SEAL”: A New Step Towards Self-Im

The concept of AI self-improvement has been a hot topic in recent research circles, with a flurry of papers emerging and prominent figures like OpenAI CEO Sam Altman weighing in on the future of self-evolving intelligent systems. Now, a new paper from MIT, titled “Self-Adapting Language Models,” introduces SEAL (Self-Adapting LLMs), a novel framework that allows large language models (LLMs) to update their own weights. This development is seen as another significant step towards the realization of truly self-evolving AI.

The research paper, published yesterday, has already ignited considerable discussion, including on Hacker News. SEAL proposes a method where an LLM can generate its own training data through “self-editing” and subsequently update its weights based on new inputs. Crucially, this self-editing process is learned via reinforcement learning, with the reward mechanism tied to the updated model’s downstream performance.

The timing of this paper is particularly notable given the recent surge in interest surrounding AI self-evolution. Earlier this month, several other research efforts garnered attention, including Sakana AI and the University of British Columbia’s “Darwin-Gödel Machine (DGM),” CMU’s “Self-Rewarding Training (SRT),” Shanghai Jiao Tong University’s “MM-UPT” framework for continuous self-improvement in multimodal large models, and the “UI-Genie” self-improvement framework from The Chinese University of Hong Kong in collaboration with vivo.

Adding to the buzz, OpenAI CEO Sam Altman recently shared his vision of a future with self-improving AI and robots in his blog post, “The Gentle Singularity.” He posited that while the initial millions of humanoid robots would need traditional manufacturing, they would then be able to “operate the entire supply chain to build more robots, which can in turn build more chip fabrication facilities, data centers, and so on.” This was quickly followed by a tweet from @VraserX, claiming an OpenAI insider revealed the company was already running recursively self-improving AI internally, a claim that sparked widespread debate about its veracity.

Regardless of the specifics of internal OpenAI developments, the MIT paper on SEAL provides concrete evidence of AI’s progression towards self-evolution.

Understanding SEAL: Self-Adapting Language Models

The core idea behind SEAL is to enable language models to improve themselves when encountering new data by generating their own synthetic data and optimizing their parameters through self-editing. The model’s training objective is to directly generate these self-edits (SEs) using data provided within the model’s context.

The generation of these self-edits is learned through reinforcement learning. The model is rewarded when the generated self-edits, once applied, lead to improved performance on the target task. Therefore, SEAL can be conceptualized as an algorithm with two nested loops: an outer reinforcement learning (RL) loop that optimizes the generation of self-edits, and an inner update loop that uses the generated self-edits to update the model via gradient descent.

This method can be viewed as an instance of meta-learning, where the focus is on how to generate effective self-edits in a meta-learning fashion.

A General Framework

SEAL operates on a single task instance (C,τ), where C is context information relevant to the task, and τ defines the downstream evaluation for assessing the model’s adaptation. For example, in a knowledge integration task, C might be a passage to be integrated into the model’s internal knowledge, and τ a set of questions about that passage.

Given C, the model generates a self-edit SE, which then updates its parameters through supervised fine-tuning: θ′←SFT(θ,SE). Reinforcement learning is used to optimize this self-edit generation: the model performs an action (generates SE), receives a reward r based on LMθ′’s performance on τ, and updates its policy to maximize the expected reward.

The researchers found that traditional online policy methods like GRPO and PPO led to unstable training. They ultimately opted for ReST^EM, a simpler, filtering-based behavioral cloning approach from a DeepMind paper. This method can be viewed as an Expectation-Maximization (EM) process, where the E-step samples candidate outputs from the current model policy, and the M-step reinforces only those samples that yield a positive reward through supervised fine-tuning.

The paper also notes that while the current implementation uses a single model to generate and learn from self-edits, these roles could be separated in a “teacher-student” setup.

Instantiating SEAL in Specific Domains

The MIT team instantiated SEAL in two specific domains: knowledge integration and few-shot learning.

• Knowledge Integration: The goal here is to effectively integrate information from articles into the model’s weights.
• Few-Shot Learning: This involves the model adapting to new tasks with very few examples.

Experimental Results

The experimental results for both few-shot learning and knowledge integration demonstrate the effectiveness of the SEAL framework.

In few-shot learning, using a Llama-3.2-1B-Instruct model, SEAL significantly improved adaptation success rates, achieving 72.5% compared to 20% for models using basic self-edits without RL training, and 0% without adaptation. While still below “Oracle TTT” (an idealized baseline), this indicates substantial progress.

For knowledge integration, using a larger Qwen2.5-7B model to integrate new facts from SQuAD articles, SEAL consistently outperformed baseline methods. Training with synthetically generated data from the base Qwen-2.5-7B model already showed notable improvements, and subsequent reinforcement learning further boosted performance. The accuracy also showed rapid improvement over external RL iterations, often surpassing setups using GPT-4.1 generated data within just two iterations.

Qualitative examples from the paper illustrate how reinforcement learning leads to the generation of more detailed self-edits, resulting in improved performance.

While promising, the researchers also acknowledge some limitations of the SEAL framework, including aspects related to catastrophic forgetting, computational overhead, and context-dependent evaluation. These are discussed in detail in the original paper.

Original Paper: https://arxiv.org/pdf/2506.10943

Project Site: https://jyopari.github.io/posts/seal

Github Repo: https://github.com/Continual-Intelligence/SEAL

The post MIT Researchers Unveil “SEAL”: A New Step Towards Self-Improving AI first appeared on Synced.

🔗 Sumber: syncedreview.com