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

📌 MAROKO133 Eksklusif ai: Researchers from PSU and Duke introduce “Multi-Agent Sys

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 Eksklusif ai: $400 exoskeleton suit built from threads and motors deli

  Exoskeleton suits have long carried a reputation for being bulky, heavy, and prohibitively expensive. For years, full-body systems capable of lifting loads or delivering lifelike VR sensations came with six-figure price tags.

  A team of researchers is now challenging that model with Kinethreads, a lightweight exosuit built from off-the-shelf components and creative fabric engineering. At less than $500, it makes muscle-assisting and haptic technology far more accessible.

  Thread-powered movement support

  Kinethreads wraps around the chest, arms, and legs like a second skin. Nylon threads run through fabric channels, connected to compact motors that act as synthetic tendons.

  When activated, the threads tighten and contract, pulling on underlying muscles to guide movement. A Raspberry Pi coordinates the system through simple Python scripts.

  The first prototypes focused on arms. A sleeve covered the bicep and forearm, with threads running from shoulder to wrist. Four motors controlled each arm: one for shoulder flexion, one for the elbow, and two for rotation and extension.

  Designers used elastic neoprene for comfort and rigid plastic guides to keep threads in place. Tension was adjusted via sliders on a laptop connected to the system.

  The design expanded to legs, which demanded broader support.

  Wider bands around thighs and calves distributed pressure evenly and reduced hotspots. Motors shifted to a belt pack for mobility, with a lithium-polymer battery supplying two hours of active use.

  This setup allowed wearers to walk freely while receiving stabilization at the hips and knees, a valuable aid for rehabilitation.

  Vibration motors placed along the threads added another layer of feedback.

  They buzzed as muscles contracted, helping the brain anticipate available assistance. Over time, wearers depended less on these cues as their natural control improved.

  Affordable haptics for VR

  Virtual reality hardware has become affordable, but realistic physical feedback remains locked behind expensive systems.

  Most consumer headsets rely only on vibration, which quickly breaks immersion when users attempt to interact with virtual objects. Kinethreads addresses that problem with a simple, low-cost approach.

  The suit weighs under five kilograms and can be put on in less than 30 seconds. Despite its small footprint, it delivers up to 120 newtons of force and vibrotactile feedback at up to 200 hertz.

  That range allows users to feel forces as subtle as resistance from a lightweight object or as strong as the jolt of an explosion.

  Ten motorized reels mounted on a lightweight vest generate the forces. Eight are arranged around the waist, while two more provide directional pull to the torso, limbs, and even the head. By reeling in or releasing strings, the system creates tension that the body interprets as weight, resistance, or acceleration.

  Researchers designed the suit with modularity in mind. Two motor sizes balance torque with cost and weight, while encoders ensure precise tracking.

  Custom PCBs, microcontrollers, and a Unity-based software stack enable real-time coordination between movement and feedback.

  Testing showed Kinethreads could simulate a wide range of scenarios, from lifting virtual objects to bracing against sudden impacts.

  Participants found the suit light, comfortable, and surprisingly easy to wear for extended sessions.

  Kinethreads demonstrates that high-quality haptics no longer need to be locked behind expensive, rigid exoskeletons.

  With smart engineering and consumer-grade materials, immersive VR and rehabilitation support can finally reach a much wider audience.

  To explore the full design, user studies, and results, check out the original research paper.

  🔗 Sumber: interestingengineering.com

  ---