MAROKO133 Update ai: Which Agent Causes Task Failures and When?Researchers from PSU and Du

📌 MAROKO133 Breaking ai: Which Agent Causes Task Failures and When?Researchers fro

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 Breaking ai: AI Backlash Grew Massively in 2025 Terbaru 2025

For the tech of the future, generative AI sure is making a lot of enemies in a very short time.

If 2023 was the year of AI’s awakening and 2024 the year of its frantic adoption, then it’s safe to say 2025 will be remembered as the year it all came crashing back down to reality. From boardrooms to classrooms, from game studios to Senate committees, this will be the year many remember as the one when AI finally wore out its welcome.

For many small towns throughout the US, the AI boom has come wafting in on a cloud of noxious smog that critics say is leading to to increased cancer risk and hostile takeovers of local infrastructure, not to mention skyrocketing electricity bills. That being the case, it’s no surprise a huge number of rural communities have spent the year organizing and agitating to shut down the tech industry’s data center projects wherever they pop up.

Indeed, from the Great Lakes to the Pacific Northwest, it seems the one thing people can agree on is that the data centers powering the AI gold rush make horrible neighbors.

Beyond the data centers, AI is helping corporations increase the rate at which they exploit workers — both on the job and at home. Back in spring, for example, Visa announced plans to let AI agents loose on customers’ financial information, one of many companies to experiment with a pivot to AI customer service.

Of course, most Americans would rather bash their heads against the wall then talk to an AI customer service agent — not that there’s much of a difference, they’d argue — as consumer sentiment surveys show. The hatred is so palpable that Americans have even taken to accusing human customer service agents of being AI when they don’t get their way.

Corporate big wigs aren’t the only ones getting fat off the AI boom, to be sure. With newfound tools to pump out ultra-realistic content at an industrial scale, Facebook scammers, art forgers, and racist influencers are also getting in on the fun.

This kind of wild west approach has fueled a rise in protest movements like Pause AI, a group calling for a halt of AI development until we can figure out what the hell is going on. 2025 even gave us our first anti-AI hunger strikes, spontaneous protests by activists in San Francisco and London. Others have risen up against AI-powered surveillance dragnets powered by companies like Flock Safety.

Amidst the outcry, a handful of politicians in the US seem to have taken notice. Vermont Senator Bernie Sanders, for example, recently kicked off a campaign calling for a pause on the “unregulated sprint to develop and deploy AI.” He’s joined in his crusade by New York representative Alexandria Ocasio-Cortez, who recently blasted Republican legislators for their attempts to pass a 10-year ban on state regulation of AI.

Though the two progressive lawmakers might not have much support from their Democratic colleagues, they’re joined by divisive right-wing figures like Georgia representative Marjorie Taylor Greene and Florida governor Ron DeSantis, who have defied their own Republican party in voicing opposition to AI initiatives.

“I’m not buying the narrative that they are trying to sell us on this,” DeSantis said at a recent roundtable on AI, according to the Washington Times.

More on AI: Professor Warns That the Wealthy Are Trying to Use AI to Seize Control of Everything

The post AI Backlash Grew Massively in 2025 appeared first on Futurism.

🔗 Sumber: futurism.com