📌 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 Hot ai: Fungi-based insulation boards that absorb CO2, resist mold

  Researchers in Germany have created a mushroom-based insulation material that could offer a greener, compostable alternative to synthetic building materials and even capture carbon along the way.

  Conducted by the Institute for Circular Economy of Bio:Polymers (ibp) at the Hof University of Applied Sciences in Germany, the project, dubbed Mycobuild, aims to build mushroom mycelium boards from lab-scale to industrial production by 2026.

  Contrary to conventional insulation materials, which often consist of synthetic or mineral materials produced with high energy intensive consumption and a poor environmental balance, the new panels are grown using fungal networks.

  “Mushroom meshes actually offer numerous advantages: they are compostable, store CO2 and require less energy to produce than conventional fossil-based insulating materials,” Robert Honke, PhD, a professor at the university and project leader, stated. “They can also be shaped flexibly and are industrially scalable.”

  Boards that store carbon

  The project consists of several crucial steps, the first of which includes preparing a substrate from locally available plant residues, such as dry straw. This serves as a breeding ground for the selected types of native fungi.

  The scientists then add the mycelium to the substrate and let it grow through the material in the desired form within a few days. As it spreads, the fungal network binds the components together into a solid composite, which is then dried and heated to inactivate the fungus and stabilize the structure.

  As per the researchers, one of the main challenges is growing the fungi under strictly controlled conditions, as even minor contamination can destroy the entire culture. This requires sterile environments and careful selection of the substrate.

  Katharina Wellmanns, PhD, a research associate at the university, explained that finding the right balance remains a great challenge. “The substrate must provide enough nutrients for the mycelium to grow optimally, but must not contain too many sugars to prevent mold growth.”

  To test the material’s flexibility, moisture absorption and thermal conductivity the team then carries out multiple extensive tests. This enables them to determine if the insulation boards meet the requirements of the construction industry. 

  Meanwhile, to eliminate moisture penetration and reduce the risk of mold, one of the main barriers to commercial adoption, the team also integrated a mineral top layer to the boards. It was developed by building materials firm Johann Bergmann GmbH & Co. KG.

  New eco-friendly insulation

  The scientists apply the coating in multiple stages and rigorously test it to ensure the material retains its properties. As the technology progresses, they may soon develop fully waterproof mycelium-based insulation. This will effectively eliminate mold issues.

  “Our tests show that the mineral top layer not only protects the material, but also increases its strength,” Wellmanns explained. “We are working on optimizing the manufacturing process to achieve complete waterproofing.”

  The team revealed that native mushrooms, such as oyster, honeydew, fox bolete or giant mushroom, are particularly promising for the project. This is due to the fact that they can grow at room temperature and do not require any additional heating or cooling measures.

  “Choosing the right type of mushroom is crucial,” Wellmanns noted, adding that while some mushrooms grow faster, others form more stable structures. “The oyster mushroom in particular has proven to be a robust candidate, as it spreads quickly and forms dense networks.”

  Selecting the right nutrient substrate plays a great role in mushroom cultivation, as excess sugar can promote unwanted germ growth. Meanwhile, Honke stated that regardless of the innovation’s potential, there are still consumer concerns.

  “Many people might be skeptical about an insulation material that is based on fungi, as they fear that this could lead to mold problems in their homes,” he said in a press release.

  Backed by Germany’s DATIpilot program, the researchers are now focused on scaling production and meeting industrial building standards. If successful, the project could not only cut emissions and waste but also reshape how we insulate our homes for good.

  🔗 Sumber: interestingengineering.com

  ---