MAROKO133 Hot ai: 7,000-year-old Chinchorro mummies represent a culture that turned grief

📌 MAROKO133 Breaking ai: 7,000-year-old Chinchorro mummies represent a culture tha

Long before the Egyptians, an ancient Chilean culture practiced an unusual form of mummification: they turned their dead into dolls.

Whenever death struck this ancient community, the Chinchorro engaged in an act of grieving that essentially became art.

According to a recent study in Cambridge University Press, this fascinating group first honored the separation inherent in death. Lead author Dr. Bernardo Arriaza proposed that they ritualistically set the body aside, then departed to gather raw materials like pigments, clay, and reeds.

Going so far as to remove the flesh and extract major organs, they used these gathered ingredients to stuff and paint the body. They placed a wig upon the head and modeled the facial features and genitals. This process, as Archaeology Mag described, was nothing short of “intensive and creative,” turning the deceased into literal works of art.

Beyond being an innovative form of mortuary practice, the study suggests this elaborate ritual likely stemmed from a high rate of infant mortality; it began as an outpouring of grief, showcasing the profound role creativity plays in processing loss.

Grief becomes art

The Chinchorro began mummifying their dead well before the Egyptians immortalized the practice. However, the Chinchorro merited their own specific term, “artificial mummification,” because they refashioned their deceased into dolls or a unique form of statuary.

As the earliest Chinchorro mummies were children from the Camarones Valley, researchers discovered that their environment contained high levels of arsenic. This toxicity affected their reproductive health, resulting in miscarriages and high infant mortality rates, Phys continues.   

Dr. Arriaza believes that this complex act of tearing bodies apart, gutting them, and putting them back together again was born from deep-seated grief. In recent years, researchers have argued that the Chinchorro weren’t practicing funerary rites, but rather “art therapy.”

“It has been a slow process of sorting through my thoughts to explain the Chinchorro’s early, complex, and creative treatment of the dead—children in particular,” Dr. Arriaza stated in Phys.

“The transformed body became a canvas for expressing emotion, and a place where these ancient people may have found emotional healing and comfort. They venerated their departed as visual icons.”

A culture marked by loss

Over time, they adopted this coping mechanism as a central rite of their culture. As Archaeology Mag stated,they symbolically kept the dead within the social world of the living, almost mimicking the “grief bots” that have made headlines as of late: AI replicas of the deceased that could almost carry on living.

“Grief is a universal emotion that connects us all,” as per the study, “…bringing people together and making social bonds stronger.” The Chinchorro suffered an unspeakable loss, as their offspring perished, which impacted how they treated the dead over the course of their existence. To weave in the enduring power of art: in processing complex emotions, a study of Andean grief suggests that these rituals resulted in particularly strong bonds between parent and child.

Like a recent archaeological discovery in Korea, which highlighted a period where that culture suffered incredible loss and destruction, the Chinchorro turned a moment of tragedy into an artistic achievement. The study of these Chinchorro mummies presents a group of people who processed their grief so profoundly that it became a work of art.

🔗 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