MAROKO133 Eksklusif ai: Scientists Intrigued by Old Drug That Reverses Signs of Alzheimer’

📌 MAROKO133 Hot ai: Scientists Intrigued by Old Drug That Reverses Signs of Alzhei

The next new treatment for Alzheimer’s disease may be an already-existing drug, according to a team of researchers in Japan.

In a series of experiments, administering an oral dose of an amino acid called arginine, which is already prescribed to treat high blood pressure, was able to suppress the buildup of a protein associated with Alzheimer’s in mice, the scientists report in a new study published in the journal Neurochemistry International.

“Our study demonstrates that arginine can suppress amyloid-beta aggregation both in vitro and in vivo,” study coauthor Yoshitaka Nagai, neuroscientist at Kindai University, said in a statement about the work. “What makes this finding exciting is that arginine is already known to be clinically safe and inexpensive, making it a highly promising candidate for repositioning as a therapeutic option for Alzheimer’s disease.”

Scientists still don’t understand the underlying cause of Alzheimer’s disease, but amyloid-beta proteins figure somewhere in the question. Though they’re a part of normal brain function, they’re sticky and can clump together to form “plaques” in the brain. These plaques are considered a hallmark of Alzheimer’s, though not all patients with the disease are found to have them.

According to the researchers’ findings, the arginine can help flush these plaques and break them apart. They demonstrated this by feeding mice with amyloid-beta buildup in their brains drinking water and food infused with small doses of the drug.

Along with breaking up the buildup, they also found that the mice which were administered arginine showed improved behavior and cognitive performance, suggesting that the drug helped reverse some of the disease’s harmful effects. This was assessed by analyzing how the mice navigated an elevated Y-shaped maze, noting how far the mice traveled and how many times it entered the maze’s “open arms,” a test of a healthy mice’s natural instinct to avoid open spaces and enter enclosed ones.

Human clinical trials are needed to bear out the medical potential, but the researchers are optimistic.

“Given its excellent safety profile and low cost, arginine could be rapidly translated to clinical trials for Alzheimer’s and potentially other related disorders,” Nagai said in the statement.

Other recent studies have explored promising avenues for treating Alzheimer’s. A team of scientists in China said they were able to almost instantly reverse the disease’s progression using nanoparticles injected in their brains that cleared the plaques and led to cognitive improvements. Another team in Japan used synthetic peptides to reverse progression in early stages of the disease.

Still, amyloid-beta’s function in the brain in general remains a mystery, and so the jury’s still out on whether targeting them is a meaningful way of treating, let alone curing, the tragic disease.

More on Alzheimer’s: Lab Mice Exposed to Microplastics Show Signs of Dementia

The post Scientists Intrigued by Old Drug That Reverses Signs of Alzheimer’s in Mice appeared first on Futurism.

🔗 Sumber: futurism.com

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

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