📌 TOPINDIATOURS Hot ai: Which Agent Causes Task Failures and When?Researchers from

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: chain.zhang@jiqizhixin.com

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

📌 TOPINDIATOURS Update ai: New solvent-based method pulls lithium faster, cleaner

Demand for lithium is rising fast as electric vehicles, wind turbines, and grid-scale batteries multiply worldwide. But extracting the metal remains slow, water-intensive, and limited to a few locations with high-quality deposits.

Most lithium today comes from salty underground brines or hard rock mining. Both routes carry heavy environmental costs.

Solar evaporation ponds consume vast amounts of land and water, while mining creates large volumes of waste and emissions.

These limits are becoming harder to ignore as battery demand accelerates. Existing extraction methods cannot scale quickly enough or reach many lithium-rich brines that contain impurities or low concentrations.

Now, researchers at Columbia Engineering say they have developed a faster and cleaner way to extract lithium directly from brines, even when concentrations are low and contamination is high.

Rethinking lithium separation

The new technique is called switchable solvent selective extraction, or S3E. It uses a temperature-sensitive solvent to pull lithium ions out of brine without relying on long evaporation cycles or binding chemicals.

At room temperature, the solvent selectively absorbs lithium along with water. When heated, the solvent releases the lithium into a purified stream and regenerates itself for reuse.

The process also removes magnesium, a common contaminant that complicates lithium recovery.

According to the study, S3E achieved lithium selectivity up to 10 times higher than sodium and 12 times higher than potassium. Magnesium was excluded through a controlled precipitation step.

“There’s no way solar evaporation alone can match future demand,” said Ngai Yin Yip, La Von Duddleson Krumb Associate Professor of Earth and Environmental Engineering at Columbia University.

“And there are promising lithium-rich brines, like those in California’s Salton Sea, where this method simply can’t be used at all.”

Nearly 40 percent of global lithium production begins with brines stored under deserts. Solar evaporation can take up to two years and requires dry climates, flat land, and heavy water use. Many brine deposits do not meet those conditions.

Faster cycles, wider reach

In lab tests using synthetic brines modeled on California’s Salton Sea, the S3E system recovered nearly 40 percent of lithium over four cycles using the same solvent batch.

The Salton Sea region is estimated to hold enough lithium to supply more than 375 million electric vehicle batteries.

The researchers say this points toward continuous operation rather than batch processing. The system can also run on low-grade heat from waste sources or solar collectors, reducing energy demands.

“This is a new way to do direct lithium extraction,” said Yip. “It’s fast, selective, and easy to scale. And it can be powered by low-grade heat from waste sources or solar collectors.”

The team stresses that the technology is still at a proof-of-concept stage. It has not yet been optimized for maximum yield or efficiency. Field testing and scale-up remain future steps.

Still, S3E could offer an alternative to evaporation ponds and hard-rock mining, opening access to lithium resources that are currently unusable.

“We talk about green energy all the time,” said Yip. “But we rarely talk about how dirty some of the supply chains are. If we want a truly sustainable transition, we need cleaner ways to get the materials it depends on. This is one step in that direction.”

The study appears in the journal Joule.

🔗 Sumber: interestingengineering.com