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

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: Carbon dioxide turned into starch: China’s new lab-base

  Researchers at the Tianjin Institute of Industrial Biotechnology have reportedly found a way to synthesize starch directly from carbon dioxide. Achieved using only enzymes and raw materials, this new process, they report, is 10 times more productive than previous attempts.

  What’s more, the process doesn’t rely on plants or photosynthesis, and could help make industrial-scale manmade starch production commercially viable.

  Starch is an important material as it is vital for the production of things like sweeteners and thickeners. It is also an essential raw material in the production of paper and textiles, pharmaceuticals, adhesives, and certain plastics.

  Today, most industrial starch is derived from corn, which requires large amounts of land to cultivate the feed crop. Owing to this, cornstarch production also consumes vast amounts of water and relies heavily on the use of fertilizers and pesticides.

  Starch from the lab, not the land

  It also has applies pressure on precious land as cornstarch production for industry is in direct competition with things like growing food crops. So, while starch is relatively cheap to make (all things considered), it is not without its compromises in terms of environmental impact.

  To help alleviate some of these issues, the researchers decided to look at a new way to produce starch synthetically in a lab. So instead of the traditional conversion of carbon dioxide to corn (and starch) using plants, they thought it should be possible to cutout the middle man, or rather plant.

  While not exactly a breakthrough (the team actually pioneered it back in 2021), they have now refined it to be close to 10x faster and more efficient.

  According to the team’s original breakthrough, this process first involves converting the carbon dioxide into methanol in the presence of an organic catalyst.

  More enzymes are then used to turn the methanol into sugars, which, in turn, are then combined to make something called polymeric starch. All told, the process requires around eleven steps to get from carbon dioxide to starch.

  The product is, according to the team, almost identical to more traditionally produced starch from corn.

  “If the overall cost of the process can be reduced to a level economically comparable with agricultural planting in the future, it is expected to save more than 90 percent of cultivated land and freshwater resources,” Yanhe Ma, a microbiologist at the Tianjin Institute of Industrial Biotechnology explained back in 2021.

  10x improvement

  Importantly, this new process makes carbon dioxide a raw material rather than a pollutant. Its use of enzymes also means that starch can now be produced without sacrificing vital farmland.

  This was achieved, the team reports, through the use of special new enzymes optimized for the task. These enzymes help slash energy costs for starch production, but also greatly improve reaction efficiency.

  The process can also be conducted under controlled conditions inside a laboratory, meaning yields are far more predictable.

  It is important to note that the 10x improvement doesn’t mean the team has managed to improve on nature. This process is not, for example, ten times better than conventional corn crop starch production.

  It just means that they’ve dramatically improved on their previous work, meaning it is one step closer to being an economically viable alternative to corn crop starch production.

  🔗 Sumber: interestingengineering.com

  ---