MAROKO133 Eksklusif ai: Researchers from PSU and Duke introduce “Multi-Agent Systems Autom

📌 MAROKO133 Hot ai: Researchers from PSU and Duke introduce “Multi-Agent Systems A

Share My Research is Synced’s column that welcomes scholars to share their own research breakthroughs with over 2M global AI enthusiasts. Beyond technological advances, Share My Research also calls for interesting stories behind the research and exciting research ideas. 

Meet the author
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

2. failure-responsible agent and the decisive error step that led to the task’s failure.

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 m…

  Konten dipersingkat otomatis.

  🔗 Sumber: syncedreview.com

  ---

  📌 MAROKO133 Update ai: The Number of Robotaxis Tesla Is Actually Running Will Make

  If Elon Musk is to be believed, millions of driverless Tesla Robotaxis are set to conquer the streets of America by next year.

  Right now, however, his tiny fleet would be stretched thin covering a couple city blocks.

  With Tesla remaining tight-lipped about how many self-driving cabs it has in operation, Texas A&M engineering student Ethan McKanna tells Electrek that he’s created an online tracker that keeps tabs on the automaker’s service in Austin, Texas by reverse-engineering its ride-hailing app. It shows that just 32 different Tesla Model Ys are currently operating as part of the Robotaxi network — and worse yet, most of the cars don’t even operate concurrently: the data suggests that fewer than ten robotaxis are giving rides at the same time.

  Taken together, it’s an embarrassing reality check to Musk’s grand promises about scaling up the program at an impossible pace.

  “This is speculative on my part, but it’s my best guess based on the little data we have and can collect,” McKanna told Electrek. “One person I talked to who scoped out the depot and recorded videos told me he believes there are 1-5 out at a time.”

  “The highly sporadic wait time shifts and my experience of consistently getting the same vehicle multiple times when I use the service in the data all corroborate that,” he added.

  McKanna’s data lines up with Tesla fans’ rough estimates of the size of the service, which launched in late June with around a dozen cars. It would also mean that Musk’s boast that it had “doubled” its fleet was technically true. 

  To gather the data, McKanna explained that he found that the Robotaxi app’s API is able to fetch ETA estimates from Tesla prior to booking a ride.

  “I have a server where every 5 minutes I ping Tesla at ~10 points in both service areas, pull the wait time, and store it,” he explained. “If a wait time is offered, I count it as available, if ‘high service demand’ or any other type of error is shown, it is marked as unavailable.”

  The tracker pings 11 different services locations within the city. A map created using the data shows that the Robotaxis are unavailable in most locations, suggesting that the “high service demand” error is actually covering up the fact that there’s little to no supply available.

  Musk has staked Tesla’s future on Robotaxis — part of his broader pivot to AI, robotics, and automation — and to support it, he’s made a number of incredible claims that are yet to be borne out, if ever. He promised that over a thousand Robotaxis would be patrolling Austin “within a few months” of launching, and that over a million autonomous Teslas would be hitting the streets in 2026 after releasing the Robotaxi software to Tesla owners. In July, he also predicted that Robotaxi operations would cover “half the population of the US by the end of the year.”

  The actual performance of the cabs would suggest they are far from being ready for primetime. For one, they still rely on the supervision of an in-car human “safety monitor,” who have already had to make several interventions. The cars have been caught violating traffic laws and have gotten into a number of accidents, the details of which Tesla has censored heavily. (Musk recently revealed that Tesla is testing letting its Robotaxis go fully driverless without any human supervision in the car.)

  In sum, it’s not a great look for the Robotaxi fleet. But Musk should know by now that size isn’t everything, right?

  More on self-driving cabs: Waymos Cause Traffic Jams Across City During Power Outage

  The post The Number of Robotaxis Tesla Is Actually Running Will Make You Snort Out of Your Nose With Pure Derision appeared first on Futurism.

  🔗 Sumber: futurism.com

  ---