📌 TOPINDIATOURS Eksklusif ai: Researchers from PSU and Duke introduce “Multi-Agent

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

  ---

  📌 TOPINDIATOURS Update ai: Chinese automaker’s 677-hp luxury off-roader with 848 m

  Dongfeng Motor’s luxury off-road brand, M-Hero, has officially hit a production milestone of 10,000 units for its flagship M817 model, signaling a rapid scale-up for the premium “intelligent” SUV.

  To mark the achievement, the brand personally delivered one of its milestone vehicles to Jin Yuzhi, CEO of Huawei’s Intelligent Automotive Solution Business Unit—a symbolic gesture underscoring the deep technical partnership between the two giants.

  The model’s commercial momentum is also extending beyond China: on December 16, the first batch of 400 M-Hero M817 units was shipped to the Middle East, marking a key step in the brand’s overseas expansion. Positioned as a large premium off-road SUV, the M-Hero M817 measures 200.8 inches long, 78.6 inches wide, and 74.5 inches high (75.7 inches with optional equipment), riding on a 118.3-inch wheelbase. 

  It offers a maximum wading depth of 2.95 feet and a tight 11.5-foot minimum turning radius. The SUV supports nine specialized driving modes, including snow, deep snow, mud, rocks, sand, and water, emphasizing its capability in extreme terrain.

  High-performance hybrid SUV with 0-62 mph in 5.2 seconds

  The M-Hero M817’s plug-in hybrid system delivers an impressive combined output of 505 kW (677 hp) and 626 lb-ft (848 Nm) of torque, allowing the SUV to accelerate from 0 to 62 mph in just 5.2 seconds. Its top-tier variant is equipped with a 50.4 kWh CATL Freevoy battery pack, providing a pure electric range of up to 134 miles under the CLTC standard. 

  For longer journeys, the comprehensive range reaches an impressive 848 miles, making it one of the most capable off-road PHEVs on the market. This combination of power, efficiency, and off-road versatility positions the M-Hero M817 as a strong contender in the global intelligent SUV segment.

  Equipped with Huawei’s Qiankun ADS 4.0, the M-Hero M817 features a comprehensive sensor suite that includes 11 high-definition cameras, three 4D millimeter-wave radars, a 192-line lidar, and 12 ultrasonic sensors, bringing the total to 27 advanced sensing units. 

  The SUV’s automatic emergency braking system can operate at speeds of up to 81 mph, providing enhanced safety across a range of driving conditions. This combination of sensors and intelligent driving capabilities positions the M-Hero M817 at the forefront of off-road vehicles, integrating novel autonomous assistance technology.

  M-Hero brand drives growth with nearly 2,000 sales in November

  In November, Dongfeng’s M-Hero brand sold 2,007 vehicles, marking a 38.4 percent increase compared with October. The sales rise also aligns with broader interest in China’s new energy vehicle segment, particularly plug-in hybrids and electric SUVs. 

  While the M-Hero remains a niche model within Dongfeng’s overall lineup, its performance suggests early traction in both domestic and international markets. With production continuing to scale, the brand is well-positioned to build on this momentum in the months ahead, potentially increasing both domestic and export deliveries. 

  Furthermore, this could help the brand strengthen its presence in China’s competitive NEV market while laying the groundwork for gradual expansion into overseas markets where demand for rugged, technology-rich SUVs is rising.

  🔗 Sumber: interestingengineering.com

  ---