MAROKO133 Hot ai: Which Agent Causes Task Failures and When?Researchers from PSU and Duke

📌 MAROKO133 Eksklusif ai: Which Agent Causes Task Failures and When?Researchers fr

Share My Research is Synced’s column that welcomes scholars to share their own research breakthroughs with over 1.5M global AI enthusiasts. Beyond technological advances, Share My Research also calls for interesting stories behind the research and exciting research ideas. Contact us: [email protected]

Meet the authors
Institutions: Penn State University, Duke University, Google DeepMind, University of Washington, Meta, Nanyang Technological University, and Oregon State University. The co-first authors are Shaokun Zhang of Penn State University and Ming Yin of Duke University.

In recent years, LLM Multi-Agent systems have garnered widespread attention for their collaborative approach to solving complex problems. However, it’s a common scenario for these systems to fail at a task despite a flurry of activity. This leaves developers with a critical question: which agent, at what point, was responsible for the failure? Sifting through vast interaction logs to pinpoint the root cause feels like finding a needle in a haystack—a time-consuming and labor-intensive effort.
 
This is a familiar frustration for developers. In increasingly complex Multi-Agent systems, failures are not only common but also incredibly difficult to diagnose due to the autonomous nature of agent collaboration and long information chains. Without a way to quickly identify the source of a failure, system iteration and optimization grind to a halt.
 
To address this challenge, researchers from Penn State University and Duke University, in collaboration with institutions including Google DeepMind, have introduced the novel research problem of “Automated Failure Attribution.” They have constructed the first benchmark dataset for this task, Who&When, and have developed and evaluated several automated attribution methods. This work not only highlights the complexity of the task but also paves a new path toward enhancing the reliability of LLM Multi-Agent systems.

The paper has been accepted as a Spotlight presentation at the top-tier machine learning conference, ICML 2025, and the code and dataset are now fully open-source.

Paper：https://arxiv.org/pdf/2505.00212
Code：https://github.com/mingyin1/Agents_Failure_Attribution
Dataset：https://huggingface.co/datasets/Kevin355/Who_and_When
 
 
Research Background and Challenges
LLM-driven Multi-Agent systems have demonstrated immense potential across many domains. However, these systems are fragile; errors by a single agent, misunderstandings between agents, or mistakes in information transmission can lead to the failure of the entire task.

Currently, when a system fails, developers are often left with manual and inefficient methods for debugging:
Manual Log Archaeology : Developers must manually review lengthy interaction logs to find the source of the problem.
Reliance on Expertise : The debugging process is highly dependent on the developer’s deep understanding of the system and the task at hand.
 
This “needle in a haystack” approach to debugging is not only inefficient but also severely hinders rapid system iteration and the improvement of system reliability. There is an urgent need for an automated, systematic method to pinpoint the cause of failures, effectively bridging the gap between “evaluation results” and “system improvement.”

Core Contributions
This paper makes several groundbreaking contributions to address the challenges above:
1. Defining a New Problem: The paper is the first to formalize “automated failure attribution” as a specific research task. This task is defined by identifying the failure-responsible agent and the decisive error step that led to the task’s failure.
2. Constructing the First Benchmark Dataset: Who&When : This dataset includes a wide range of failure logs collected from 127 LLM Multi-Agent systems, which were either algorithmically generated or hand-crafted by experts to ensure realism and diversity. Each failure log is accompanied by fine-grained human annotations for:
Who: The agent responsible for the failure.
When: The specific interaction step where the decisive error occurred.
Why: A natural language explanation of the cause of the failure.

3. Exploring Initial “Automated Attribution” Methods : Using the Who&When dataset, the paper designs and assesses three distinct methods for automated failure attribution:
– All-at-Once: This method provides the LLM with the user query and the complete failure log, asking it to identify the responsible agent and the decisive error step in a single pass. While cost-effective, it may struggle to pinpoint precise errors in long contexts.
– Step-by-Step: This approach mimics manual debugging by having the LLM review the interaction log sequentially, making a judgment at each step until the error is found. It is more precise at locating the error step but incurs higher costs and risks accumulating errors.
– Binary Search: A compromise between the first two methods, this strategy repeatedly divides the log in half, using the LLM to determine which segment contains the error. It then recursively searches the identified segment, offering a balance of cost and performance.

Experimental Results and Key Findings 
Experiments were conducted in two settings: one where the LLM knows the ground truth answer to the problem the Multi-Agent system is trying to solve (With Ground Truth) and one where it does not (Without Ground Truth). The primary model used was GPT-4o, though other models were also tested. The systematic evaluation of these methods on the Who&When dataset yielded several important insights:
– A Long Way to Go: Current methods are far from perfect. Even the best-performing single method achieved an accuracy of only about 53.5% in identifying the responsible agent and a mere 14.2% in pinpointing the exact error step. Some methods performed even worse than random guessing, underscoring the difficulty of the task.
– No “All-in-One” Solution: Different methods excel at different aspects of the problem. The All-at-Once method is better at identifying “Who,” while the Step-by-Step method is more effective at determining “When.” The Binary Search method provides a middle-ground performance.
 

– Hybrid Approaches Show Promise but at a High Cost: The researchers found that combining different methods, such as using the All-at-Once approach to identify a potential agent and then applying the Step-by-Step method to find the error, can improve overall performance. However, this comes with a significant increase in computational cost.

– State-of-the-Art Models Struggle: Surprisingly, even the most advanced reasoning models, like OpenAI o1 and DeepSeek R1, find this task challenging.- This h…

Konten dipersingkat otomatis.

🔗 Sumber: syncedreview.com

📌 MAROKO133 Update ai: Top African Health Official Blasts Trump Administration’s P

Last week, officials at the Africa Centres for Disease Control and Prevention (Africa CDC) announced that health authorities in Guinea-Bissau had moved to halt a controversial study which would have used unvaccinated infants as lab rats. The US Department of Health and Human Services (HHS), however, insisted the trial was still on.

That wasn’t all US health authorities had to say. In a bizarre outburst, the HHS told Futurism that the Africa CDC was a “fake and powerless organization,” and insisted that it was “not a reliable source” for information. During a press meeting on Thursday, director-general of the Africa CDC Jean Kaseya hit back.

“Let me tell you: we are not an NGO,” Kaseya said defiantly. “We are not a UN organization. We are the convening power in Africa. The mandate [is] given to us by all 55 African heads of state and government.”

“People can meet in the US, people can meet in Europe,” Kaseya continued, likely referencing leaked emails between HHS officials and Danish researchers to push the vaccine trial in Guinea-Bissau. “If we are not there, they are wasting their time.”

Kaseya also made it clear that no government has the right to impose experiments on the people of Guinea-Bissau.

“Africa CDC is respecting and supporting the sovereignty of the country,” the official continued. “It’s not Africa CDC that will say, ‘this clinical trial will take place or not.’ It’s not any other international body that will come to say, ‘this clinical trial will take place or not.’ It’s not a foreign country that will come and say, ‘this one will take place.’ It’s the sovereignty of the country.”

Though the HHS evidently thinks it has the power to push dangerous vaccines experiments on African countries, the reality isn’t so simple. According to the Africa CDC’s guidelines, any vaccine trial that doesn’t receive a written authorization from the country’s National Medicines Regulatory Authority is illegal. Likewise, it must have approval from the National Ethics Committee, as well as a local institutional review board at each site where the trial is to take place.

On top of all that, any trial must also have explicit approval from the country’s Ministry of Health — granting officials in Guinea-Bissau the power to functionally veto any experiment the government doesn’t support.

Kaseya emphasized the importance of exercising sovereignty over African public health. “Our vision is not coming from Western countries,” he said. “Our vision is coming from Africa, shaped by African leadership, based on African realities.”

In an interview with the Guardian, Abdulhammad Babatunde, a doctor and global health researcher in Nigeria, said that it’s “very important to fund research that Africans actually want.”

“Africans want to solve Africa’s problems, not satisfy the curiosity of the funders,” Babatunde said, referring to the HHS, which announced the $1.6 million funding package for the study in December.

Once up and running, the vaccine trial would have given 7,000 newborn children a vaccine for hepatitis B, while 7,000 more infants in the control group wouldn’t receive the vaccine.

As the Guardian notes, nearly one in five adults and 11 percent of children in Guinea-Bissau have hepatitis B — which carries a significant risk for debilitating illness and death. In such a health environment, there’s no justification for purposefully leaving infants unvaccinated for any length of time.

“This is not acceptable,” Babatunde told the paper. “To prevent things like the Tuskegee study and others, the control group has to get the standard of care, and the intervention group should get [potentially] better care.”

Asked whether the HHS’s comments would affect future collaboration with Africa CDC, Kaseya said he’s putting the strange episode behind him.

“I was briefed that they don’t know anything about any statement against Africa CDC,” he said. “I could tell you about a statement that could be somewhere, made by someone. But officials from HHS just yesterday — senior people — said they don’t know anything about the statement. I trust them, I’ve closed the chapter.”

More on Africa: AI “Companion Bots” Actually Run by Exploited Kenyans, Worker Claims

The post Top African Health Official Blasts Trump Administration’s Plans for Human Experimentation in Africa appeared first on Futurism.

🔗 Sumber: futurism.com